Will AI replace Engineers, All Other?

Not as a job, but it is already doing parts of the work. Across the 179 official task statements scored for Engineers, All Other (United States, SOC 17-2199), 32% of the importance-weighted core work is made of tasks today's AI could already do most of. The overall exposure score is 46 out of 100 (range 40–52, band: partial). That is a statement about tasks, not about headcount: this measures what AI could do, not whether any employer adopts it, whether the law allows it, or whether doing the routine parts faster creates more demand for the human parts. Figures are from release 2026-q4.1.

Which tasks in “Engineers, All Other” can AI already do?

The highest-scoring tasks in release 2026-q4.1 are: “Maintain technical project files” (93/100, very high); “Develop customer documentation, such as performance specifications, training manuals, or operating instructions” (93/100, very high); “Make system device lists or event timing charts” (93/100, very high). Each score comes from five published 0–4 ratings turned into a number by a published formula, and each carries the model's one-sentence reason on the page.

Which tasks in “Engineers, All Other” stay human?

About 45% of this job's task weight sits in work that scores low for AI exposure. The lowest-scoring tasks in release 2026-q4.1 are: “Synthesize, process, or characterize nanomaterials, using advanced tools or techniques” (0/100, minimal); “Install, calibrate, operate, or maintain robots” (0/100, minimal); “Direct balance of plant (BOP) construction, generator installation, testing, commissioning, or supervisory control and data acquisition (SCADA) to ensure com…” (6/100, minimal). Low scores usually mean the task needs a body in a room, a legally accountable human, or trust built in real time. Those are the three things the scoring rubric treats as gates rather than obstacles.

What should someone working in “Engineers, All Other” do about AI?

Start from the ledger rather than the headline: 32% of this job's weighted core work is exposed, and roughly 45% is not. The practical move is to spend more of your week on the tasks that score low, the ones above, and to get fluent at directing AI through the tasks that score high, because those are the parts that change whether or not you are ready for them. This page does not predict your job, and nothing here is career advice tailored to you: the score describes the occupation, not the person.

How is the AI exposure score for Engineers, All Other calculated?

Each official task statement for the occupation is rated on five published 0–4 dimensions (output replicability, physical embodiment, licensed accountability, real-time human trust, and data availability) by claude-opus-5 using scoring prompt task_scoring_v1.0. The model never writes the score; a published formula turns the five ratings into a 0–100 number, so every score can be recomputed by hand. The occupation figure is the importance-weighted mean across 179 scored tasks. The prompt, the rubric, the formula and the full dataset are published at https://futureproof.collab365.com/method and https://futureproof.collab365.com/data/2026-q4.1 under CC BY 4.0.

US dataswitch to UK

Engineers, All Other

identifying and recommending energy savings strategies to achieve more energy-efficient operation, investigating mechanical failures or unexpected maintenance problems and conducting or overseeing the conduct of prototype development or microfabrication activities to ensure compliance to specifications and promote effective production processes. If that's your week, this page is about your job.

The honest answer

AI is already taking a real slice of the routine work here: processing or interpreting signals or sensor data. That is a slice of tasks, not of you.

Your move: what you can actually do about this ↓

Where the work sits, by task weight

shifting to AI: 32%
changing shape: 23%
staying human: 45%

These bars are tasks changing hands, not people being counted out. The ledger below shows which.

Whole-job exposure score 46 out of 100 (40–52 allowing for uncertainty): partial exposure, across 179 scored tasks. The number is the support for the sentence above it, not a headline about anyone’s future.

How we know this

What is measured: Every published task statement for engineers, all other is rated on five dimensions: can a model produce the output, does the work need a body in a room, does it need a legally accountable person, does it depend on a person being trusted in the moment, and how much data exists. A published formula turns those five ratings into the score; the model never writes the number.

How the bar is built: Each task’s share of the bar is its published importance weight, so a task you do all day counts for more than one you do twice a year.

Release: 2026-q4.1, scores computed 2026-08-04. Read the full method.

That slice is not coming back; the core of the job, synthesizing, processing or characterize nanomaterials, using advanced tools or techniques, stays yours. The tools change, the responsibility doesn't.

Your week, as this page understands it

All engineers not listed separately. The job title says “engineers” or “all other”: officially one job, two names. The real job is the part underneath: synthesizing, processing or characterize nanomaterials, using advanced tools or techniques. That is the thing someone has to be right about.

The exposed part of this job is specific, and we won’t pretend it is coming back. But engineers, all other is not one task. It is 179 scored ones, and they are nowhere near equally exposed. The most durable of them, on this evidence, is synthesizing, processing or characterize nanomaterials, using advanced tools or techniques, and the ledger below shows exactly why.

Where the work sits, by task weight

shifting to AI: 32%
changing shape: 23%
staying human: 45%

These bars are tasks changing hands, not people being counted out. The ledger below shows which.

How we know this

How the bar is built: Each task’s share of the bar is its published importance weight, so a task you do all day counts for more than one you do twice a year.

Release: 2026-q4.1, scores computed 2026-08-04. Read the full method.

Your job, task by task

These are the official task statements for this occupation, in plain English, sorted by what the evidence says is happening to each one. The official wording sits under every line so you can check the rewrite against it.

One task scored differently between repeat runs, so its range on this page is wider. We would rather show the wobble than hide it.
19 of this occupation's scored task statements carry a score that was measured under a different occupation's context, because the statement is shared between jobs and has only been scored once so far. Each one names the occupation it was measured under in the free routes below; none is presented as a measurement for this job.
Task clusters are not derived in this release, so the task-cluster field is empty and no Collab365 Space routing is attached to this occupation yet.

Shifting to AI

50 tasks

Tasks today’s tools can already do most of. This is the part we will not soften: where these rows are the bulk of your week, the week changes.

Creating plans for solar energy system development
This is reading one thing and writing another: plans in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Create plans for solar energy system development, monitoring, and evaluation activities.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Development and monitoring plans follow standard formats, so good drafts are achievable with engineer review.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Identifying and recommending energy savings strategies to achieve more energy-efficient operation
This is reading one thing and writing another: energy savings strategies in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Identify and recommend energy savings strategies to achieve more energy-efficient operation.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (68–82 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Spotting savings from consumption data and standard measures suits software, though the best options depend on the building.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Performing computer simulation of solar photovoltaic
This is reading one thing and writing another: computer simulation of solar photovoltaic in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Perform computer simulation of solar photovoltaic (PV) generation system performance or energy production to optimize efficiency.” (O*NET task statement)
How this row was scored
Exposure score: 83 out of 100 (79–87 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Solar production modelling uses well-documented software and public weather data, so it automates well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Reviewing or approving designs, calculations or cost estimates
This is reading one thing and writing another: designs, calculations or cost estimates in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Review or approve designs, calculations, or cost estimates.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Checking designs, calculations and estimates is document work AI does well, but the approval carries an engineer’s responsibility.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.

Changing shape

40 tasks

Tasks where the machine takes the producing and a person keeps the checking, the deciding, or the answering-for-it. For most jobs this is the biggest group, and it is where "transformation, not termination" is literally visible.

Creating electrical single-line diagrams, panel schedules or connection diagrams for solar electric systems
The software now makes the first pass at electrical single-line diagrams, panel schedules or connection diagrams, but someone has to be answerable for the result, and it cannot be the software. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Create electrical single-line diagrams, panel schedules, or connection diagrams for solar electric systems, using computer-aided design (CAD) software.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; someone qualified has to answer for it.
The rating behind it: Single-line diagrams and panel schedules are highly standardized, but a licensed engineer must produce them for permitting.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Monitoring energy related design or construction issues
The software now makes the first pass at energy related design, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Monitor energy related design or construction issues, such as energy engineering, energy management, or sustainable design.” (O*NET task statement)
How this row was scored
Exposure score: 42 out of 100 (35–49 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Design and construction issues can be tracked from documents, but keeping a project honest usually means site visits and conversations.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 1/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Conducting research related to a range of nanotechnology topics
The software now makes the first pass at research, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Conduct research related to a range of nanotechnology topics, such as packaging, heat transfer, fluorescence detection, nanoparticle dispersion, hybrid systems, liquid systems, nanocomposites, nanofabrication, optoelectronics, or nanolithography.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Published nanotechnology research is plentiful, but original work still means experiments in a lab.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.

Staying human

89 tasks

Tasks that stay with a person, because they happen in the physical world, because the rules need someone accountable, or because the value is that a specific person does them.

Conducting engineering site audits to collect structural
This work happens in the physical world: site audits, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Conduct engineering site audits to collect structural, electrical, and related site information for use in the design of residential or commercial solar power systems.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: A site audit means measuring roofs, panels and electrical gear in person.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Supervising technologists or technicians engaged in nanotechnology research or production
The value here is that a specific person handles technologists and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Supervise technologists or technicians engaged in nanotechnology research or production.” (O*NET task statement)
How this row was scored
Exposure score: 13 out of 100 (9–17 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the value is that a specific person does it.
The rating behind it: Supervising researchers and technicians depends on working with the people directly.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 3/4 · how much data exists 2/4.
Providing technical direction or supporting to installation teams during installation
This work happens in the physical world: technical direction, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Provide technical direction or support to installation teams during installation, start-up, testing, system commissioning, or performance monitoring.” (O*NET task statement)
How this row was scored
Exposure score: 20 out of 100 (13–27 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; work that happens in the physical world; the value is that a specific person does it.
The rating behind it: Installation support means being available on site and guiding the crew directly.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.

Show the other 169 tasks

Maintaining technical project files
shifting to AI
This is reading one thing and writing another: technical project files in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Maintain technical project files.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Keeping project files organized and current is straightforward document management.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing customer documentation, such as performance specifications, training manuals or operating instructions
shifting to AI
This is reading one thing and writing another: customer documentation in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Develop customer documentation, such as performance specifications, training manuals, or operating instructions.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Manuals, specifications and operating instructions are standard technical writing, which AI produces to a high standard.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Making system device lists or event timing charts
shifting to AI
This is reading one thing and writing another: system device lists in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Make system device lists or event timing charts.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Device lists and timing charts are structured documents generated straight from system data.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Processing or interpreting signals or sensor data
shifting to AI
This is reading one thing and writing another: signals in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Process or interpret signals or sensor data.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Turning sensor signals into useful information is a well-established computing task.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Analyzing operation of wind farms or wind farm components to determine reliability
shifting to AI
This is reading one thing and writing another: operation of wind farms in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Analyze operation of wind farms or wind farm components to determine reliability, performance, and compliance with specifications.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Turbine performance data streams in continuously, and analyzing it for reliability and compliance suits software well.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Creating checklists for review or inspection of completed solar installation projects
shifting to AI
This is reading one thing and writing another: checklists in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Create checklists for review or inspection of completed solar installation projects.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Inspection checklists are standard documents built from published requirements.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Analyzing, interpreting or creating graphical representations of energy data, using engineering software
shifting to AI
This is reading one thing and writing another: graphical representations of energy data in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Analyze, interpret, or create graphical representations of energy data, using engineering software.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Turning energy data into charts and interpretations is exactly the kind of analysis software does routinely.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Creating or maintaining photonic design histories
shifting to AI
This is reading one thing and writing another: photonic design histories in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Create or maintain photonic design histories.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Keeping a design history file complete and current is straightforward record work.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Documenting photonics system or component design processes
shifting to AI
This is reading one thing and writing another: photonics system in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Document photonics system or component design processes, including objectives, issues, or outcomes.” (O*NET task statement)
How this row was scored
Exposure score: 93 out of 100 (89–97 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Recording design decisions, issues and outcomes is document work software handles reliably.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Creating embedded software design programs
shifting to AI
This is reading one thing and writing another: embedded software design programs in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Create embedded software design programs.” (O*NET task statement)
How this row was scored
Exposure score: 83 out of 100 (79–87 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Embedded control code is a well-documented kind of programming that AI drafts capably.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Writing algorithms or programming code for ad hoc robotic applications
shifting to AI
This is reading one thing and writing another: algorithms in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Write algorithms or programming code for ad hoc robotic applications.” (O*NET task statement)
How this row was scored
Exposure score: 83 out of 100 (79–87 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Writing code for robotic applications is one of the strongest current uses of these tools.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Researching renewable or alternative energy systems or technologies
shifting to AI
This is reading one thing and writing another: renewable in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Research renewable or alternative energy systems or technologies, such as solar thermal or photovoltaic energy.” (O*NET task statement)
How this row was scored
Exposure score: 83 out of 100 (79–87 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Researching renewable technologies means reading and summarizing published material, which AI does very well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Publishing engineering reports documenting design details or qualification test results
shifting to AI
This is reading one thing and writing another: reports documenting design details in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Publish engineering reports documenting design details or qualification test results.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (77–85 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Engineering reports written from design and test data are structured documents software drafts well.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Documenting robotic application development, maintenance or changes
shifting to AI
This is reading one thing and writing another: robotic application development, maintenance or changes in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Document robotic application development, maintenance, or changes.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (77–85 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Writing up development and maintenance changes is structured documentation software drafts well.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Writing reports to document wind farm collector system test results
shifting to AI
This is reading one thing and writing another: reports in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Write reports to document wind farm collector system test results.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (77–85 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Writing up test results into a report is document work, though an engineer stands behind the conclusions.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Preparing energy-related project reports or related documentation
shifting to AI
This is reading one thing and writing another: energy-related project reports in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Prepare energy-related project reports or related documentation.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (77–85 allowing for uncertainty): very high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Writing up project findings into reports is document work, though an engineer puts their name to the conclusions.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing formal documentation for microelectromechanical systems
shifting to AI
This is reading one thing and writing another: formal documentation in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Develop formal documentation for microelectromechanical systems (MEMS) devices, including quality assurance guidance, quality control protocols, process control checklists, data collection, or reporting.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (71–79 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Quality documents, checklists and reporting formats are highly standardized writing tasks.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Identifying new applications for existing nanotechnologies
shifting to AI
This is reading one thing and writing another: new applications in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Identify new applications for existing nanotechnologies.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (71–79 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Spotting new uses for an existing technology is largely a matter of reading widely and connecting ideas.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Monitoring and analyzing energy consumption
shifting to AI
This is reading one thing and writing another: energy consumption in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Monitor and analyze energy consumption.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (71–79 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Consumption data is digital and the patterns are well understood, so tracking and analyzing it is largely automatic.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Determining commercial, industrial, scientific or other uses for electro-optical applications or devices
shifting to AI
This is reading one thing and writing another: commercial, industrial, scientific or other uses in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Determine commercial, industrial, scientific, or other uses for electro-optical applications or devices.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (68–82 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Scanning markets and literature for uses of a technology is research work AI does well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Analyzing system performance or operational requirements
shifting to AI
This is reading one thing and writing another: system performance in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Analyze system performance or operational requirements.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (68–82 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Performance data and requirements documents are exactly the kind of material software analyzes well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Considering environmental issues when proposing product designs involving microelectromechanical systems
shifting to AI
This is reading one thing and writing another: environmental issues in, a record out. That is the shape today's tools are built for.
importance not published
O*NET hasn't yet published importance ratings for this occupation's tasks, so the ordering here is ours, not theirs.
Source: “Consider environmental issues when proposing product designs involving microelectromechanical systems (MEMS) technology.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (71–79 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Environmental impact considerations are documented in standards and literature, so drafting this assessment is straightforward.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Recommending best fuel for specific sites or circumstances
shifting to AI
This is reading one thing and writing another: best fuel in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Recommend best fuel for specific sites or circumstances.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (68–82 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Fuel choice follows documented costs and emissions, though local supply and site quirks shape the final recommendation.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Writing reports or proposals related to photonics research or development projects
shifting to AI
This is reading one thing and writing another: reports in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Write reports or proposals related to photonics research or development projects.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (71–79 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Project reports and proposals are writing tasks AI drafts well once the technical findings exist.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Determining the feasibility, costs or performance benefits of new mechatronic equipment
shifting to AI
This is reading one thing and writing another: the feasibility in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Determine the feasibility, costs, or performance benefits of new mechatronic equipment.” (O*NET task statement)
How this row was scored
Exposure score: 75 out of 100 (71–79 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Feasibility and cost comparisons rest on documented prices and performance figures, which is straightforward analysis.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing or filing intellectual property and patent disclosure or application documents related to microelectromechanical systems
shifting to AI
This is reading one thing and writing another: intellectual property in, a record out. That is the shape today's tools are built for.
importance 4 · Supplemental
Source: “Develop or file intellectual property and patent disclosure or application documents related to microelectromechanical systems (MEMS) devices, products, or systems.” (O*NET task statement)
How this row was scored
Exposure score: 72 out of 100 (65–79 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Patent and disclosure documents follow strict, well-published formats, though a qualified attorney handles the filing.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Preparing nanotechnology-related invention disclosures or patent applications
shifting to AI
This is reading one thing and writing another: nanotechnology-related invention disclosures in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Prepare nanotechnology-related invention disclosures or patent applications.” (O*NET task statement)
How this row was scored
Exposure score: 72 out of 100 (65–79 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Invention disclosures and patent drafts follow strict published formats, though filings go through a qualified attorney.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Developing standard operation procedures and quality or safety standards for solar installation work
shifting to AI
This is reading one thing and writing another: standard operation procedures in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Develop standard operation procedures and quality or safety standards for solar installation work.” (O*NET task statement)
How this row was scored
Exposure score: 72 out of 100 (65–79 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Safety and quality procedures are published in codes and standards, making them straightforward to draft.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Promoting awareness or use of alternative or renewable energy sources
shifting to AI
This is reading one thing and writing another: awareness in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Promote awareness or use of alternative or renewable energy sources.” (O*NET task statement)
How this row was scored
Exposure score: 70 out of 100 (63–77 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Awareness material is easy to produce, but persuading people to change how they use energy works better from a person.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 4/4.
Creating back-ups of robot programs or parameters
shifting to AI
This is reading one thing and writing another: back-ups of robot programs in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Create back-ups of robot programs or parameters.” (O*NET task statement)
How this row was scored
Exposure score: 69 out of 100 (65–73 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Backing up programs and parameters is routine, largely automatic work, though someone has to connect to the controller.
The five ratings: output a model can produce 4/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Verifying energy bills and meter readings
shifting to AI
This is reading one thing and writing another: energy bills in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Verify energy bills and meter readings.” (O*NET task statement)
How this row was scored
Exposure score: 69 out of 100 (65–73 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Checking bills against meter readings is simple arithmetic on digital records, though some meters still have to be read on site.
The five ratings: output a model can produce 4/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Creating or maintaining formal engineering documents
shifting to AI
This is reading one thing and writing another: formal engineering documents in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Create or maintain formal engineering documents, such as schematics, bills of materials, components or materials specifications, or packaging requirements.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Schematics, parts lists and specifications follow set formats, so drafts come easily and an engineer checks them.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Recommending process or infrastructure changes to improve wind turbine performance
shifting to AI
This is reading one thing and writing another: process in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Recommend process or infrastructure changes to improve wind turbine performance, reduce operational costs, or comply with regulations.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Performance data is digital and the improvement options are documented, so recommendations can be drafted well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing specifications for wind technology components
shifting to AI
This is reading one thing and writing another: specifications in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Develop specifications for wind technology components, such as gearboxes, blades, generators, frequency converters, or pad transformers.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Component specifications follow established formats and standards, so strong drafts are achievable.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Reviewing specifications and recommending engineering or manufacturing changes to achieve solar design objectives
shifting to AI
This is reading one thing and writing another: specifications in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Review specifications and recommend engineering or manufacturing changes to achieve solar design objectives.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Reviewing specifications against design goals is document work AI drafts well, with an engineer deciding.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Performing thermal, stress or cost reduction analyses for solar systems
shifting to AI
This is reading one thing and writing another: thermal, stress or cost reduction analyses in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Perform thermal, stress, or cost reduction analyses for solar systems.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Thermal, stress and cost calculations follow documented methods and are well suited to software.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing design specifications and functional requirements
shifting to AI
This is reading one thing and writing another: design specifications in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Develop design specifications and functional requirements for residential, commercial, or industrial solar energy systems or components.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: System specifications follow established templates and standards, so strong drafts are achievable.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Reviewing architectural, mechanical or electrical plans or specifications to evaluate energy efficiency
shifting to AI
This is reading one thing and writing another: architectural, mechanical or electrical plans or specifications in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Review architectural, mechanical, or electrical plans or specifications to evaluate energy efficiency.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (59–73 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Plans and specifications are documents, and checking them against efficiency standards is work AI can draft well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Identifying materials appropriate for mechatronic system designs
shifting to AI
This is reading one thing and writing another: this work in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Identify materials appropriate for mechatronic system designs.” (O*NET task statement)
How this row was scored
Exposure score: 66 out of 100 (62–70 allowing for uncertainty): high exposure, high confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Identifying suitable materials is comparison against documented property and cost data.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing software to control robotic systems
shifting to AI
It is the same call made over and over on software, with a right answer to check it against. That is what a model is trained on.
importance 4 · Core
Source: “Design software to control robotic systems for applications, such as military defense or manufacturing.” (O*NET task statement)
How this row was scored
Exposure score: 65 out of 100 (58–72 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Robot control code is well documented, but safety-critical software needs an engineer to design and prove it.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Planning or scheduling engineering research or development projects involving microelectromechanical systems
shifting to AI
This is reading one thing and writing another: research in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Plan or schedule engineering research or development projects involving microelectromechanical systems (MEMS) technology.” (O*NET task statement)
How this row was scored
Exposure score: 64 out of 100 (57–71 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Project plans and schedules follow known formats, though agreeing them involves the research team.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Communicating operating characteristics or performance experience to other engineers or designers for training or new product development purposes
shifting to AI
This is reading one thing and writing another: characteristics in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Communicate operating characteristics or performance experience to other engineers or designers for training or new product development purposes.” (O*NET task statement)
How this row was scored
Exposure score: 64 out of 100 (57–71 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Performance findings can be written up automatically, but passing on hard-won experience works best person to person.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Preparing reports, deliver presentations or participating in program review activities to communicate engineering results or recommendations
shifting to AI
This is reading one thing and writing another: reports, deliver presentations in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Prepare reports, deliver presentations, or participate in program review activities to communicate engineering results or recommendations.” (O*NET task statement)
How this row was scored
Exposure score: 64 out of 100 (57–71 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Reports and slides are easy to draft, but presenting and defending results is done by a person.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Training personnel or clients on topics
shifting to AI
This is reading one thing and writing another: personnel in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Train personnel or clients on topics such as energy management.” (O*NET task statement)
How this row was scored
Exposure score: 64 out of 100 (57–71 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Training material is easy to produce, though people often prefer a live session so they can ask questions.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Training operators, engineers or other personnel
shifting to AI
This is reading one thing and writing another: operators, engineers or other personnel in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Train operators, engineers, or other personnel.” (O*NET task statement)
How this row was scored
Exposure score: 64 out of 100 (57–71 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Training material is easy to produce, though hands-on subjects usually need a person in the room.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Conducting research on new photonics technologies
shifting to AI
This is reading one thing and writing another: research in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Conduct research on new photonics technologies.” (O*NET task statement)
How this row was scored
Exposure score: 62 out of 100 (55–69 allowing for uncertainty): high exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Published photonics research is abundant and AI summarizes it well, though real research also happens at the bench.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 4/4.
Proposing product designs involving microelectromechanical systems
changing shape
The software now makes the first pass at product designs involving microelectromechanical systems, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Core
Source: “Propose product designs involving microelectromechanical systems (MEMS) technology, considering market data or customer requirements.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Turning market and customer information into a product proposal is research and writing software supports, though engineers judge what is buildable.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Conducting analyses addressing issues, failure, reliability or yield improvement
changing shape
The software now makes the first pass at analyses addressing issues, failure, reliability or yield improvement, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Conduct analyses addressing issues such as failure, reliability, or yield improvement.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Failure and yield data can be crunched automatically, but explaining why a microdevice fails needs an expert.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing automated robotic systems to increase production volume or precision in high-throughput operations
changing shape
The software now makes the first pass at automated robotic systems, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Design automated robotic systems to increase production volume or precision in high-throughput operations, such as automated ribonucleic acid (RNA) analysis or sorting, moving, or stacking production materials.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: High-throughput automation designs follow known patterns, but each application needs engineering judgment.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing robotic systems, such as automatic vehicle control, autonomous vehicles, advanced displays, advanced sensing, robotic platforms, computer vision or telematics systems
changing shape
The software now makes the first pass at robotic systems, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Design robotic systems, such as automatic vehicle control, autonomous vehicles, advanced displays, advanced sensing, robotic platforms, computer vision, or telematics systems.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: AI can draft parts of a robotic system design, but the whole architecture needs an experienced engineer.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing active control algorithms, electronics, software, electromechanical or electrohydraulic systems for wind turbines
changing shape
The software now makes the first pass at active control algorithms, electronics, software, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Develop active control algorithms, electronics, software, electromechanical, or electrohydraulic systems for wind turbines.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Turbine control software follows known principles, but safety and load behavior need engineering design.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Consulting with construction or renovation clients or other engineers on topics
changing shape
The software now makes the first pass at construction, but the part that matters is a person saying it and standing behind it. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Consult with construction or renovation clients or other engineers on topics such as Leadership in Energy and Environmental Design (LEED) or Green Buildings.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Green building standards are published in detail, but clients want a person to talk the options through with them.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 4/4.
Creating mechanical models to simulate mechatronic design concepts
changing shape
The software now makes the first pass at mechanical models, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Create mechanical models to simulate mechatronic design concepts.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (51–65 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Simulation setups can be drafted from documented methods, but building a model you can trust needs engineering judgment.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Analyzing existing development or manufacturing procedures and suggest improvements
changing shape
The software now makes the first pass at development, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Analyze existing development or manufacturing procedures and suggest improvements.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Process data can be analyzed for improvements, though understanding why a line behaves as it does often needs a walk round.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Researching, selecting or applying sensors, communication technologies or controlling devices for motion control, positioning sensing, pressure sensing or electronic communication
changing shape
The software now makes the first pass at sensors, communication technologies or controlling devices, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Research, select, or apply sensors, communication technologies, or control devices for motion control, position sensing, pressure sensing, or electronic communication.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Sensor and controller datasheets are public, so selection is easy, but fitting the parts into real hardware needs hands.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Debuging robotics programs
changing shape
The software now makes the first pass at robotics programs, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Debug robotics programs.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: AI is good at finding faults in code, though robot bugs often only show up on the real machine.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Writing or installing energy management routines for building automation systems
changing shape
The software now makes the first pass at energy management routines, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Core
Source: “Write or install energy management routines for building automation systems.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Building control routines are code, which AI can draft, but installing them on site equipment needs an engineer.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Providing consultation or training on topics
changing shape
The software now makes the first pass at consultation, but the part that matters is a person saying it and standing behind it. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Provide consultation or training on topics such as mechatronics or automated control.” (O*NET task statement)
How this row was scored
Exposure score: 53 out of 100 (46–60 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; the value is that a specific person does it.
The rating behind it: The subject matter is documented, but training and consulting land better when a person answers questions live.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Providing technical guidance or supporting to customers on topics
changing shape
The software now makes the first pass at technical guidance, but the part that matters is a person saying it and standing behind it. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Provide technical guidance or support to customers on topics such as nanosystem start-up, maintenance, or use.” (O*NET task statement)
How this row was scored
Exposure score: 53 out of 100 (46–60 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Support content is documented, but customers rely on a trusted specialist when a system misbehaves.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Providing scientific or technical guidance or expertise
changing shape
The software now makes the first pass at scientific, but the part that matters is a person saying it and standing behind it. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Provide scientific or technical guidance or expertise to scientists, engineers, technologists, technicians, or others, using knowledge of chemical, analytical, or biological processes as applied to micro and nanoscale systems.” (O*NET task statement)
How this row was scored
Exposure score: 53 out of 100 (46–60 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; the value is that a specific person does it.
The rating behind it: The science is documented, but colleagues take guidance partly on the standing of the person giving it.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Reading current literature, talk with colleagues, continue education or participating in professional organizations or conferences to keep abreast of developments in the field
changing shape
The software now makes the first pass at current literature, talk, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Read current literature, talk with colleagues, continue education, or participate in professional organizations or conferences to keep abreast of developments in the field.” (O*NET task statement)
How this row was scored
Exposure score: 53 out of 100 (46–60 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Keeping up with the field is mostly reading, with some events attended in person.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 4/4.
Advising clients or colleagues on topics
changing shape
The software now makes the first pass at clients, but the part that matters is a person saying it and standing behind it. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Advise clients or colleagues on topics such as climate control systems, energy modeling, data logging, sustainable design, or energy auditing.” (O*NET task statement)
How this row was scored
Exposure score: 53 out of 100 (46–60 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; the value is that a specific person does it.
The rating behind it: The technical content is well documented, but clients weigh advice partly on the adviser they trust.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Creating mechanical design documents
changing shape
The software now makes the first pass at mechanical design documents, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Create mechanical design documents for parts, assemblies, or finished products.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: AI can help with drawings and part documents, but a workable mechanical design still needs an engineer to shape it.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing advanced electronic control systems for mechanical systems
changing shape
The software now makes the first pass at advanced electronic control systems, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Design advanced electronic control systems for mechanical systems.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Control system design is supported by modelling tools, with the engineer owning the safety decisions.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Investigating characteristics, cost, performance or processing capability of potential microelectromechanical systems
changing shape
The software now makes the first pass at characteristics, cost, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Investigate characteristics such as cost, performance, or process capability of potential microelectromechanical systems (MEMS) device designs, using simulation or modeling software.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Simulation tools can compare design options, but interpreting them for a novel device needs a specialist.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Evaluating materials, fabrication methods, joining methods, surface treatments or packaging to ensure acceptable processing, performance, cost, sustainability or availability
changing shape
The software now makes the first pass at materials, fabrication methods, joining methods, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Supplemental
Source: “Evaluate materials, fabrication methods, joining methods, surface treatments, or packaging to ensure acceptable processing, performance, cost, sustainability, or availability.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Published data helps compare materials and processes, but the deciding details for a specific device are in-house knowledge.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Creating or maintaining wind farm layouts
changing shape
The software now makes the first pass at wind farm layouts, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Create or maintain wind farm layouts, schematics, or other visual documentation for wind farms.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Layout drawings can be part-generated, but the site-specific engineering behind them needs a qualified person.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Creating models to optimize the layout of wind farm access roads
changing shape
The software now makes the first pass at models, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Create models to optimize the layout of wind farm access roads, crane pads, crane paths, collection systems, substations, switchyards, or transmission lines.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Layout optimization suits computer modelling, but the terrain and grid constraints need engineering judgment.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Determining applications of photonics appropriate to meet product objectives or features
changing shape
The software now makes the first pass at this work, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Determine applications of photonics appropriate to meet product objectives or features.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Matching photonics options to product needs is desk analysis, though the decisive detail is specific to the product.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing or developing automated control systems for environmental applications
changing shape
The software now makes the first pass at automated control systems, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Design or develop automated control systems for environmental applications, such as waste processing, air quality, or water quality systems.” (O*NET task statement)
How this row was scored
Exposure score: 50 out of 100 (43–57 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Environmental control systems follow known patterns, but each plant’s process needs engineering judgment.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Writing proposals to secure external funding or to partner with other companies
changing shape
The software now makes the first pass at proposals, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Write proposals to secure external funding or to partner with other companies.” (O*NET task statement)
How this row was scored
Exposure score: 49 out of 100 (42–56 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: AI drafts proposals well, but a convincing funding bid depends on the team’s own track record and relationships.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Providing engineering technical support to designers of prototype wind turbines
changing shape
The software now makes the first pass at technical support, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Supplemental
Source: “Provide engineering technical support to designers of prototype wind turbines.” (O*NET task statement)
How this row was scored
Exposure score: 49 out of 100 (42–56 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Technical support can be documented, but prototype teams rely on an engineer who knows the machine.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Providing technical support for robotic systems
changing shape
The software now makes the first pass at technical support, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Provide technical support for robotic systems.” (O*NET task statement)
How this row was scored
Exposure score: 48 out of 100 (41–55 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Support answers are documented, but tricky robot problems still need someone at the machine talking to operators.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Designing, developing or implementing control circuits or algorithms for electromechanical or pneumatic devices or systems
changing shape
The software now makes the first pass at control circuits, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Design, develop, or implement control circuits or algorithms for electromechanical or pneumatic devices or systems.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Control algorithms can be drafted from textbooks, though getting circuits working involves the actual hardware.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing or validating product-specific test protocols
changing shape
The software now makes the first pass at product-specific test protocols, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Develop or validate product-specific test protocols, acceptance thresholds, or inspection tools for quality control testing or performance measurement.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Test protocols follow known patterns, but setting thresholds for a new device needs engineering judgment and trial runs.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Validating fabrication processes for microelectromechanical systems
changing shape
The software now makes the first pass at fabrication processes, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Validate fabrication processes for microelectromechanical systems (MEMS), using statistical process control implementation, virtual process simulations, data mining, or life testing.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Statistical process work suits software, but validating a fabrication line involves the line itself.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Conducting research on robotic technology to create new robotic systems or system capabilities
changing shape
The software now makes the first pass at research, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Conduct research on robotic technology to create new robotic systems or system capabilities.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Literature can be surveyed automatically, but creating genuinely new robot capabilities means building and testing them.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Conducting research into the feasibility
changing shape
The software now makes the first pass at research, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Conduct research into the feasibility, design, operation, or performance of robotic mechanisms, components, or systems, such as planetary rovers, multiple mobile robots, reconfigurable robots, or man-machine interactions.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Feasibility studies can draw on published research, but new robot mechanisms have to be built and tried.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Investigating experimental wind turbines or wind turbine technologies
changing shape
The software now makes the first pass at experimental wind turbines, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Supplemental
Source: “Investigate experimental wind turbines or wind turbine technologies for properties such as aerodynamics, production, noise, and load.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Turbine data can be analyzed, but investigating aerodynamics, noise and loads involves real machines.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Performing root cause analysis on wind turbine tower component failures
changing shape
The software now makes the first pass at root cause analysis, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Perform root cause analysis on wind turbine tower component failures.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Failure data helps, but root cause work on a tower usually involves inspecting the structure.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing underground or overhead wind farm collector systems
changing shape
The software now makes the first pass at underground, but someone has to be answerable for the result, and it cannot be the software. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Design underground or overhead wind farm collector systems.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; someone qualified has to answer for it.
The rating behind it: Collector system design follows electrical standards, but a licensed engineer must produce and stamp the design.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing or coordinating design of photovoltaic
changing shape
The software now makes the first pass at design of photovoltaic, but someone has to be answerable for the result, and it cannot be the software. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Design or coordinate design of photovoltaic (PV) or solar thermal systems, including system components, for residential and commercial buildings.” (O*NET task statement)
How this row was scored
Exposure score: 43 out of 100 (36–50 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; someone qualified has to answer for it.
The rating behind it: Solar system design follows known rules, but permit drawings must be produced by a licensed engineer.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Coordinating or supervising the work of suppliers or vendors in the designing
changing shape
The software now makes the first pass at the work of suppliers, but the part that matters is a person saying it and standing behind it. So the job becomes checking and deciding rather than producing.
importance 3 · Core
Source: “Coordinate or supervise the work of suppliers or vendors in the designing, building, or testing of nanosystem devices, such as lenses or probes.” (O*NET task statement)
How this row was scored
Exposure score: 40 out of 100 (33–47 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Supplier coordination is partly paperwork, but holding vendors to a design depends on the working relationship.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Designing engineering systems for the automation of industrial tasks
staying human
The ratings behind this row put systems well outside what today's tools can do on their own.
importance 4 · Core
Source: “Design engineering systems for the automation of industrial tasks.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Automation designs can be drafted from standard patterns, but the real plant layout and constraints need an engineer on site.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Upgrading the design of existing devices by adding mechatronic elements
staying human
The ratings behind this row put the design of existing devices well outside what today's tools can do on their own.
importance 4 · Core
Source: “Upgrade the design of existing devices by adding mechatronic elements.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Adding electronics to an existing device needs knowledge of that specific product and hands on the hardware.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Planning mobile robot paths and teaching path plans to robots
staying human
This work happens in the physical world: mobile robot paths, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Plan mobile robot paths and teach path plans to robots.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Paths can be planned in software, but teaching them to a robot usually happens at the machine.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing processes or identifying equipment needed for pilot or commercial nanoscale scale production
staying human
The ratings behind this row put processes well outside what today's tools can do on their own.
importance 4 · Core
Source: “Develop processes or identify equipment needed for pilot or commercial nanoscale scale production.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Scaling a nanoscale process to production depends on in-house process knowledge and trial runs.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Inspecting or monitoring energy systems
staying human
This work happens in the physical world: energy systems, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Inspect or monitor energy systems, including heating, ventilating, and air conditioning (HVAC) or daylighting systems to determine energy use or potential energy savings.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Meter data can be read remotely, but judging how heating, ventilation or daylighting really performs usually means being in the building.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Monitoring or calibrating automated systems
staying human
This work happens in the physical world: automated systems, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Monitor or calibrate automated systems, industrial control systems, or system components to maximize efficiency of production.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: System data can be monitored remotely, but calibration means adjusting real equipment.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing or programing robotics systems for environmental clean-up applications to minimize human exposure to toxic or hazardous materials or to improve the quality or speed of clean-up operations
staying human
The ratings behind this row put robotics systems well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design or program robotics systems for environmental clean-up applications to minimize human exposure to toxic or hazardous materials or to improve the quality or speed of clean-up operations.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Clean-up robots are specialized designs, and the site conditions matter more than published knowledge.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing robotics applications for manufacturers of green products
staying human
The ratings behind this row put robotics applications well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design robotics applications for manufacturers of green products, such as wind turbines or solar panels, to increase production time, eliminate waste, or reduce costs.” (O*NET task statement)
How this row was scored
Exposure score: 38 out of 100 (31–45 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Robot cells for green manufacturing follow known patterns but need tailoring to each factory.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Identifying procure or developing test equipment, instrumentation or facilities for characterization of microelectromechanical systems, MEMS) applications
staying human
The ratings behind this row put procure well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Identify, procure, or develop test equipment, instrumentation, or facilities for characterization of microelectromechanical systems (MEMS) applications.” (O*NET task statement)
How this row was scored
Exposure score: 37 out of 100 (30–44 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Equipment can be researched and specified from catalogues, but choosing and installing lab instruments needs presence.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Designing mechatronics components for computer-controlled products
staying human
The ratings behind this row put mechatronics components well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design mechatronics components for computer-controlled products, such as cameras, video recorders, automobiles, or airplanes.” (O*NET task statement)
How this row was scored
Exposure score: 35 out of 100 (28–42 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over.
The rating behind it: Designing components for cameras or vehicles is original product engineering that AI cannot finish on its own.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Reviewing or negotiating energy purchase agreements
staying human
The value here is that a specific person handles energy purchase agreements and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Review or negotiate energy purchase agreements.” (O*NET task statement)
How this row was scored
Exposure score: 35 out of 100 (28–42 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Contract terms can be analyzed, but negotiating a supply deal turns on bargaining with a real counterparty.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Refining final microelectromechanical systems, MEMS) design to optimize design for target dimensions, physical tolerances or processing constraints
staying human
The ratings behind this row put final microelectromechanical systems well outside what today's tools can do on their own.
importance 4 · Supplemental
Source: “Refine final microelectromechanical systems (MEMS) design to optimize design for target dimensions, physical tolerances, or processing constraints.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Squeezing a microdevice design into real tolerances and process limits is specialist work AI cannot finish.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Creating schematics and physical layouts of integrated microelectromechanical systems
staying human
The ratings behind this row put schematics well outside what today's tools can do on their own.
importance 4 · Core
Source: “Create schematics and physical layouts of integrated microelectromechanical systems (MEMS) components or packaged assemblies consistent with process, functional, or package constraints.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Microsystem layouts are highly specialized designs, and AI can only suggest fragments an engineer must redo.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing or engineering nanomaterials, nanodevices, nano-enabled products or nanosystems, using three-dimensional computer-aided design, CAD) software
staying human
The ratings behind this row put nanomaterials, nanodevices, nano-enabled products or nanosystems well outside what today's tools can do on their own.
importance 3 · Core
Source: “Design or engineer nanomaterials, nanodevices, nano-enabled products, or nanosystems, using three-dimensional computer-aided design (CAD) software.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing nanomaterials and nanodevices in three-dimensional software is specialist engineering AI cannot complete.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Performing energy modeling, measurement, verification, commissioning or retro-commissioning
staying human
This work happens in the physical world: energy modeling, measurement, verification, commissioning, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Perform energy modeling, measurement, verification, commissioning, or retro-commissioning.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Energy models can be built at a computer, but commissioning means checking that real equipment works as designed.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing electro-optical sensing or imaging systems
staying human
The ratings behind this row put electro-optical well outside what today's tools can do on their own.
importance 4 · Core
Source: “Design electro-optical sensing or imaging systems.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Electro-optical sensor design is specialist original engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing laser machining equipment
staying human
The ratings behind this row put laser machining equipment well outside what today's tools can do on their own.
importance 2 · Supplemental
Source: “Design laser machining equipment for purposes such as high-speed ablation.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing high-speed laser machining equipment is original mechanical and optical engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing gas lasers, solid state lasers, infrared or other light emitting or light sensitive devices
staying human
The ratings behind this row put gas lasers, solid state lasers well outside what today's tools can do on their own.
importance 3 · Core
Source: “Design gas lasers, solid state lasers, infrared, or other light emitting or light sensitive devices.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Laser design is deep specialist engineering where AI contributes ideas at best.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing self-monitoring mechanical systems
staying human
The ratings behind this row put mechanical systems well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design self-monitoring mechanical systems, such as gear systems that monitor loading or condition of systems to detect and prevent failures.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing systems that watch their own condition is original engineering with limited published precedent.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing or redesigning optical fibers to minimize energy loss
staying human
The ratings behind this row put optical fibers well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design or redesign optical fibers to minimize energy loss.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Redesigning optical fiber to cut losses is specialist materials engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing photonics products, such as light sources, displays or photovoltaics, to achieve increased energy efficiency
staying human
The ratings behind this row put photonics products well outside what today's tools can do on their own.
importance 3 · Core
Source: “Design photonics products, such as light sources, displays, or photovoltaics, to achieve increased energy efficiency.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing more efficient light sources or solar panels is frontier engineering, not something AI can deliver.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing solar energy photonics or other materials or devices to generate energy
staying human
The ratings behind this row put solar energy photonics well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design solar energy photonics or other materials or devices to generate energy.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing materials and devices that generate energy is original research engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing nano-enabled products with reduced toxicity
staying human
The ratings behind this row put nano-enabled products well outside what today's tools can do on their own.
importance 3 · Core
Source: “Design nano-enabled products with reduced toxicity, increased durability, or improved energy efficiency.” (O*NET task statement)
How this row was scored
Exposure score: 33 out of 100 (26–40 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing safer, longer-lasting nano-enabled products is original materials engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Overseeing the work of contractors in accordance with project requirements
staying human
The value here is that a specific person handles the work of contractors and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Oversee the work of contractors in accordance with project requirements.” (O*NET task statement)
How this row was scored
Exposure score: 30 out of 100 (23–37 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Contractor oversight means holding people to their commitments, which depends on someone being there.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Managing new product introduction projects to ensure effective deployment of microelectromechanical systems
staying human
The value here is that a specific person handles new product introduction projects and stands behind it. That is earned, not computed.
importance 3 · Supplemental
Source: “Manage new product introduction projects to ensure effective deployment of microelectromechanical systems (MEMS) devices or applications.” (O*NET task statement)
How this row was scored
Exposure score: 30 out of 100 (23–37 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Product launch plans can be drafted, but coordinating teams and suppliers depends on a person driving it.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Overseeing the work activities of wind farm consultants or subcontractors
staying human
The value here is that a specific person handles the work activities of wind farm consultants and stands behind it. That is earned, not computed.
importance 3 · Supplemental
Source: “Oversee the work activities of wind farm consultants or subcontractors.” (O*NET task statement)
How this row was scored
Exposure score: 30 out of 100 (23–37 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Overseeing consultants and subcontractors depends on holding people to their commitments.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Directing the implementation of energy management projects
staying human
The value here is that a specific person handles the implementation of energy management projects and stands behind it. That is earned, not computed.
importance 3 · Core
Source: “Direct the implementation of energy management projects.” (O*NET task statement)
How this row was scored
Exposure score: 30 out of 100 (23–37 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Directing a project means pushing people and contractors to act, which needs a person in charge.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Applying mechatronic or automated solutions to the transfer of materials
staying human
This work happens in the physical world: mechatronic, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Apply mechatronic or automated solutions to the transfer of materials, components, or finished goods.” (O*NET task statement)
How this row was scored
Exposure score: 29 out of 100 (22–36 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Material handling designs can be drafted, but making them work involves the actual factory floor.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Automating assays on laboratory robotics
staying human
This work happens in the physical world: assays, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Automate assays on laboratory robotics.” (O*NET task statement)
How this row was scored
Exposure score: 29 out of 100 (22–36 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Assay scripts can be written on screen, but setting the robot up on the bench is hands-on.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing or conducting tests of new nanotechnology products
staying human
This work happens in the physical world: tests of new nanotechnology products, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Design or conduct tests of new nanotechnology products, processes, or systems.” (O*NET task statement)
How this row was scored
Exposure score: 29 out of 100 (22–36 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Test designs can be drafted, but running them on real nanomaterials means lab work.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Evaluating robotic systems or prototypes
staying human
This work happens in the physical world: robotic systems, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Evaluate robotic systems or prototypes.” (O*NET task statement)
How this row was scored
Exposure score: 29 out of 100 (22–36 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Evaluating a robot or prototype means watching it run, though the analysis afterwards is desk work.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing advanced precision equipment for accurate or controlled applications
staying human
The ratings behind this row put advanced precision equipment well outside what today's tools can do on their own.
importance 4 · Core
Source: “Design advanced precision equipment for accurate or controlled applications.” (O*NET task statement)
How this row was scored
Exposure score: 28 out of 100 (21–35 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing precision equipment is original engineering where AI offers ideas rather than a usable design.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing or developing vacuum tube collector systems for solar applications
staying human
The ratings behind this row put vacuum tube collector systems well outside what today's tools can do on their own.
importance 2 · Supplemental
Source: “Design or develop vacuum tube collector systems for solar applications.” (O*NET task statement)
How this row was scored
Exposure score: 28 out of 100 (21–35 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Vacuum tube collector design is specialist thermal engineering AI cannot complete.
The five ratings: output a model can produce 1/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Developing laser-processed designs, such as laser-cut medical devices
staying human
The rules require a named, qualified person to answer for laser-processed designs, and that person cannot be a piece of software.
importance 3 · Supplemental
Source: “Develop laser-processed designs, such as laser-cut medical devices.” (O*NET task statement)
How this row was scored
Exposure score: 28 out of 100 (18–38 allowing for uncertainty): low exposure, medium confidence, and it moved between repeat runs, so the range is widened.
Why it sits in this group: someone qualified has to answer for it.
The rating behind it: Designs can be drafted, but medical device work needs a named engineer accountable for the result.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Managing the development, design or construction of energy conservation projects to ensure acceptability of budgets and time lines
staying human
The value here is that a specific person handles the development and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Manage the development, design, or construction of energy conservation projects to ensure acceptability of budgets and time lines, conformance to federal and state laws, or adherence to approved specifications.” (O*NET task statement)
How this row was scored
Exposure score: 26 out of 100 (19–33 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; the value is that a specific person does it.
The rating behind it: Budgets and schedules can be drafted automatically, but running a project depends on directing people and answering to the client.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 1/4 · needs to be trusted in the moment 2/4 · how much data exists 3/4.
Developing or validating specialized materials characterization procedures
staying human
This work happens in the physical world: specialized materials characterization procedures, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Develop or validate specialized materials characterization procedures, such as thermal withstand, fatigue, notch sensitivity, abrasion, or hardness tests.” (O*NET task statement)
How this row was scored
Exposure score: 25 out of 100 (18–32 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Standard test methods are documented, but developing and proving new ones means running real samples.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Investigating mechanical failures or unexpected maintenance problems
staying human
This work happens in the physical world: mechanical failures, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Investigate mechanical failures or unexpected maintenance problems.” (O*NET task statement)
How this row was scored
Exposure score: 25 out of 100 (18–32 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Working out why something broke usually means getting hands on the failed part as well as reading data.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Monitoring wind farm construction to ensure compliance with regulatory standards or environmental requirements
staying human
This work happens in the physical world: wind farm construction, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Monitor wind farm construction to ensure compliance with regulatory standards or environmental requirements.” (O*NET task statement)
How this row was scored
Exposure score: 25 out of 100 (18–32 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Compliance paperwork is desk work, but monitoring construction means being on the wind farm.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing, integrating or testing photonics systems or components
staying human
This work happens in the physical world: photonics systems, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Design, integrate, or test photonics systems or components.” (O*NET task statement)
How this row was scored
Exposure score: 25 out of 100 (18–32 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Some design can be done at a computer, but integrating and testing optical systems means aligning real components.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Conducting energy audits to evaluate energy use and to identify conservation and cost reduction measures
staying human
This work happens in the physical world: energy audits, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Conduct energy audits to evaluate energy use and to identify conservation and cost reduction measures.” (O*NET task statement)
How this row was scored
Exposure score: 25 out of 100 (18–32 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: The analysis and report are desk work, but the audit itself starts with a survey of the building.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing electronic, mechanical or computerized processes to perform tasks in dangerous situations
staying human
The ratings behind this row put electronic, mechanical or computerized processes well outside what today's tools can do on their own.
importance 4 · Supplemental
Source: “Develop electronic, mechanical, or computerized processes to perform tasks in dangerous situations, such as underwater exploration or extraterrestrial mining.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing equipment for underwater or space work is frontier engineering with very little to copy from.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Conducting acceptance tests, vendor-qualification protocols, surveys, audits, corrective-action reviews or performance monitoring of incoming materials or components to ensure conformance
staying human
This work happens in the physical world: acceptance tests, vendor-qualification protocols, surveys, audits, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Conduct acceptance tests, vendor-qualification protocols, surveys, audits, corrective-action reviews, or performance monitoring of incoming materials or components to ensure conformance to specifications.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Paperwork for acceptance testing is easy to produce, but audits and incoming checks involve inspecting real materials.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Devising microelectromechanical systems, MEMS) production methods, such as integrated circuit fabrication, lithographic electroform modeling or micromachining
staying human
The ratings behind this row put microelectromechanical systems, mems) production methods well outside what today's tools can do on their own.
importance 4 · Supplemental
Source: “Devise microelectromechanical systems (MEMS) production methods, such as integrated circuit fabrication, lithographic electroform modeling, or micromachining.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Inventing a fabrication route for a new microdevice is frontier work with little published precedent.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing end-of-arm tooling
staying human
The ratings behind this row put end-of-arm well outside what today's tools can do on their own.
importance 4 · Core
Source: “Design end-of-arm tooling.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Gripper design draws on catalogue parts, though each application needs bespoke engineering and trials.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Engineering production processes for specific nanotechnology applications
staying human
The ratings behind this row put production processes well outside what today's tools can do on their own.
importance 3 · Core
Source: “Engineer production processes for specific nanotechnology applications, such as electroplating, nanofabrication, or epoxy.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Engineering a production process for a specific nanotechnology is original work grounded in the plant itself.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Developing optical or imaging systems
staying human
The ratings behind this row put optical well outside what today's tools can do on their own.
importance 4 · Core
Source: “Develop optical or imaging systems, such as optical imaging products, optical components, image processes, signal process technologies, or optical systems.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Developing optical and imaging systems is original engineering that needs lab work to prove out.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Overseeing or providing expertise on manufacturing
staying human
This work happens in the physical world: expertise, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Oversee or provide expertise on manufacturing, assembly, or fabrication processes.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Manufacturing expertise is partly documented, but overseeing fabrication means being on the floor with the team.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Designing or developing energy products using nanomaterials or nanoprocesses
staying human
The ratings behind this row put energy products well outside what today's tools can do on their own.
importance not published
O*NET hasn't yet published importance ratings for this occupation's tasks, so the ordering here is ours, not theirs.
Source: “Design or develop energy products using nanomaterials or nanoprocesses, such as micro-nano machining.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Building energy devices from nanomaterials is research-stage engineering with very little to copy.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing or developing industrial air quality microsystems
staying human
The ratings behind this row put industrial air quality microsystems well outside what today's tools can do on their own.
importance not published
O*NET hasn't yet published importance ratings for this occupation's tasks, so the ordering here is ours, not theirs.
Source: “Design or develop industrial air quality microsystems, such as carbon dioxide fixing devices.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Air quality microsystems are experimental devices that need original laboratory engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing or developing sensors to reduce the energy or resource requirements to operate appliances
staying human
The ratings behind this row put sensors well outside what today's tools can do on their own.
importance not published
O*NET hasn't yet published importance ratings for this occupation's tasks, so the ordering here is ours, not theirs.
Source: “Design or develop sensors to reduce the energy or resource requirements to operate appliances, such as washing machines or dishwashing machines.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing low-power appliance sensors is original hardware work AI can only assist with.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing sensors or switches that require little or no power to operate for environmental monitoring or industrial metering applications
staying human
The ratings behind this row put sensors well outside what today's tools can do on their own.
importance not published
O*NET hasn't yet published importance ratings for this occupation's tasks, so the ordering here is ours, not theirs.
Source: “Design sensors or switches that require little or no power to operate for environmental monitoring or industrial metering applications.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Sensors that run on almost no power are frontier hardware designs requiring lab development.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Researching or developing emerging microelectromechanical
staying human
The ratings behind this row put microelectromechanical well outside what today's tools can do on their own.
importance not published
O*NET hasn't yet published importance ratings for this occupation's tasks, so the ordering here is ours, not theirs.
Source: “Research or develop emerging microelectromechanical (MEMS) systems to convert nontraditional energy sources into power, such as ambient energy harvesters that convert environmental vibrations into usable energy.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Energy harvesting microsystems are early research, so AI can suggest directions but not deliver a design.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Developing photonics sensing or manufacturing technologies to improve the efficiency of manufacturing or related processes
staying human
The ratings behind this row put photonics well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Develop photonics sensing or manufacturing technologies to improve the efficiency of manufacturing or related processes.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: New sensing and manufacturing technologies come from lab development, not from existing knowledge alone.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Applying nanotechnology to improve the performance or reduce the environmental impact of energy products
staying human
The ratings behind this row put nanotechnology well outside what today's tools can do on their own.
importance 3 · Core
Source: “Apply nanotechnology to improve the performance or reduce the environmental impact of energy products, such as fuel cells or solar cells.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Improving fuel cells or solar cells with nanotechnology is research-stage engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing nano-based manufacturing processes to minimize water
staying human
The ratings behind this row put nano-based manufacturing processes well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design nano-based manufacturing processes to minimize water, chemical, or energy use, as well as to reduce waste production.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Cutting water, chemical and energy use in a nanoscale process needs original process engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing nanoparticle catalysts to detect or remove chemical or other pollutants from water
staying human
The ratings behind this row put nanoparticle catalysts well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design nanoparticle catalysts to detect or remove chemical or other pollutants from water, soil, or air.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Designing catalysts to strip pollutants from water or air is experimental chemistry.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Designing nanosystems with components, nanocatalysts or nanofiltration devices to clean specific pollutants from hazardous waste sites
staying human
The ratings behind this row put nanosystems well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Design nanosystems with components such as nanocatalysts or nanofiltration devices to clean specific pollutants from hazardous waste sites.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Pollutant-cleaning nanosystems are experimental designs proven in the lab, not on paper.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Integrating nanotechnology with antimicrobial properties into products
staying human
The ratings behind this row put nanotechnology well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Integrate nanotechnology with antimicrobial properties into products, such as household or medical appliances, to reduce the development of bacteria or other microbes.” (O*NET task statement)
How this row was scored
Exposure score: 24 out of 100 (17–31 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the five ratings behind the score, with no single dominant reason.
The rating behind it: Adding antimicrobial nanotechnology to products is research-stage materials work.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Demonstrating miniaturized systems that contain components
staying human
This work happens in the physical world: miniaturized systems, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Demonstrate miniaturized systems that contain components, such as microsensors, microactuators, or integrated electronic circuits, fabricated on silicon or silicon carbide wafers.” (O*NET task statement)
How this row was scored
Exposure score: 21 out of 100 (14–28 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Demonstrating a working miniature system means showing real hardware to real people.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Conducting or overseeing the conduct of prototype development or microfabrication activities to ensure compliance to specifications and promote effective production processes
staying human
This work happens in the physical world: the conduct of prototype development, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Conduct or oversee the conduct of prototype development or microfabrication activities to ensure compliance to specifications and promote effective production processes.” (O*NET task statement)
How this row was scored
Exposure score: 21 out of 100 (14–28 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Prototype runs happen in a cleanroom, so overseeing them means being there as well as reading the data.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Assisting in the transition of photonic prototypes to production
staying human
This work happens in the physical world: the transition of photonic prototypes, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Assist in the transition of photonic prototypes to production.” (O*NET task statement)
How this row was scored
Exposure score: 21 out of 100 (14–28 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Moving a prototype to production involves the factory and the people running it.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Collecting data for energy conservation analyses
staying human
This work happens in the physical world: data, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Collect data for energy conservation analyses, using jobsite observation, field inspections, or sub-metering.” (O*NET task statement)
How this row was scored
Exposure score: 19 out of 100 (15–23 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Collecting the data means being on the jobsite, walking the building and fitting sub-meters.
The five ratings: output a model can produce 3/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Developing or implementing microelectromechanical systems
staying human
This work happens in the physical world: microelectromechanical systems, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Develop or implement microelectromechanical systems (MEMS) processing tools, fixtures, gages, dies, molds, or trays.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Making tooling, fixtures and trays for microfabrication is specialist hands-on work.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Conducting experimental or virtual studies to investigate characteristics and processing principles of potential microelectromechanical systems
staying human
This work happens in the physical world: experimental, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Conduct experimental or virtual studies to investigate characteristics and processing principles of potential microelectromechanical systems (MEMS) technology.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Investigating how a new micro-technology behaves needs laboratory experiments; software can support the thinking but not the bench work.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Creating designs or prototypes for nanosystem applications
staying human
This work happens in the physical world: designs, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Create designs or prototypes for nanosystem applications, such as biomedical delivery systems or atomic force microscopes.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Prototyping nanoscale devices is specialist laboratory engineering.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Developing catalysis or other green chemistry methods to synthesize nanomaterials
staying human
This work happens in the physical world: catalysis, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Develop catalysis or other green chemistry methods to synthesize nanomaterials, such as nanotubes, nanocrystals, nanorods, or nanowires.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Developing greener ways to make nanomaterials means chemistry at the bench.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Developing green building nanocoatings, such as self-cleaning, anti-stain, depolluting, anti-fogging, anti-icing, antimicrobial, moisture-resistant or ultraviolet protectant coatings
staying human
This work happens in the physical world: green building nanocoatings, in a real place. Software cannot follow it there.
importance 2 · Supplemental
Source: “Develop green building nanocoatings, such as self-cleaning, anti-stain, depolluting, anti-fogging, anti-icing, antimicrobial, moisture-resistant, or ultraviolet protectant coatings.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Developing new coatings means making and testing them in a lab.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Reengineering nanomaterials to improve biodegradability
staying human
This work happens in the physical world: nanomaterials, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Reengineer nanomaterials to improve biodegradability.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Making nanomaterials biodegradable is laboratory chemistry, not a paper exercise.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Implementing or testing design solutions
staying human
This work happens in the physical world: design solutions, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Implement or test design solutions.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Testing a design means building it and putting it on the bench.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Integrating robotics with peripherals, such as welders, controllers or other equipment
staying human
This work happens in the physical world: robotics, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Integrate robotics with peripherals, such as welders, controllers, or other equipment.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Wiring a robot to welders and controllers is hands-on integration work.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Testing wind turbine components, using mechanical or electronic testing equipment
staying human
This work happens in the physical world: wind turbine components, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Test wind turbine components, using mechanical or electronic testing equipment.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Testing components means putting real hardware on test equipment.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Testing or evaluating photovoltaic, PV) cells or modules
staying human
This work happens in the physical world: photovoltaic, pv) cells or modules, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Test or evaluate photovoltaic (PV) cells or modules.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Testing solar cells and modules means measuring real hardware.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Analyzing fabricate or testing fiber-optic links
staying human
This work happens in the physical world: fabricate, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Analyze, fabricate, or test fiber-optic links.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Fiber optic links have to be built and measured physically.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Conducting testing to determine functionality or optimization or to establish limits of photonics systems or components
staying human
This work happens in the physical world: determine functionality, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Conduct testing to determine functionality or optimization or to establish limits of photonics systems or components.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Determining a system’s real limits means measuring actual hardware in the lab.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Building configure or testing robots or robotic applications
staying human
This work happens in the physical world: configure, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Build, configure, or test robots or robotic applications.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (10–18 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Building and testing a robot means assembling and running real hardware.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Conducting harsh environmental testing, accelerated aging, device characterization or field trials to validate devices
staying human
This work happens in the physical world: harsh environmental testing, accelerated aging, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Conduct harsh environmental testing, accelerated aging, device characterization, or field trials to validate devices, using inspection tools, testing protocols, peripheral instrumentation, or modeling and simulation software.” (O*NET task statement)
How this row was scored
Exposure score: 13 out of 100 (9–17 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Environmental and ageing tests mean putting real devices into real chambers and instruments.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Supervising technologists, technicians or other engineers
staying human
The value here is that a specific person handles technologists, technicians or other engineers and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Supervise technologists, technicians, or other engineers.” (O*NET task statement)
How this row was scored
Exposure score: 13 out of 100 (9–17 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the value is that a specific person does it.
The rating behind it: Supervising engineers and technicians depends on a working relationship with the people being led.
The five ratings: output a model can produce 1/4 · needs a body in a room 1/4 · needs an accountable person 0/4 · needs to be trusted in the moment 3/4 · how much data exists 2/4.
Designing or developing new crystals for photonics applications
staying human
This work happens in the physical world: new crystals, in a real place. Software cannot follow it there.
importance 2 · Supplemental
Source: “Design or develop new crystals for photonics applications.” (O*NET task statement)
How this row was scored
Exposure score: 13 out of 100 (9–17 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Developing new crystals is laboratory work with materials, and much of the know-how is unpublished.
The five ratings: output a model can produce 1/4 · needs a body in a room 2/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Generating high-resolution images or measuring force-distance curves
staying human
This work happens in the physical world: high-resolution images, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Generate high-resolution images or measure force-distance curves, using techniques such as atomic force microscopy.” (O*NET task statement)
How this row was scored
Exposure score: 8 out of 100 (4–12 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Atomic force microscopy means operating a delicate instrument on a real sample.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Testing wind turbine equipment to determine effects of stress or fatigue
staying human
This work happens in the physical world: wind turbine equipment, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Test wind turbine equipment to determine effects of stress or fatigue.” (O*NET task statement)
How this row was scored
Exposure score: 8 out of 100 (4–12 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Stress and fatigue testing needs the equipment physically loaded and measured.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Selecting, purchasing, setting up, operating or troubleshooting state-of-the-art laser cutting equipment
staying human
This work happens in the physical world: up, operating or troubleshooting state-of-the-art laser cutting equipment, in a real place. Software cannot follow it there.
importance 2 · Supplemental
Source: “Select, purchase, set up, operate, or troubleshoot state-of-the-art laser cutting equipment.” (O*NET task statement)
How this row was scored
Exposure score: 8 out of 100 (4–12 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Setting up, running and fixing laser cutting equipment means hands on the machine.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Developing or testing photonic prototypes or models
staying human
This work happens in the physical world: photonic prototypes, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Develop or test photonic prototypes or models.” (O*NET task statement)
How this row was scored
Exposure score: 8 out of 100 (4–12 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Building and testing photonic prototypes means working with real optics on a bench.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Overseeing operation of microelectromechanical systems
staying human
This work happens in the physical world: operation of microelectromechanical systems, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Oversee operation of microelectromechanical systems (MEMS) fabrication or assembly equipment, such as handling, singulation, assembly, wire-bonding, soldering, or package sealing.” (O*NET task statement)
How this row was scored
Exposure score: 7 out of 100 (3–11 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Overseeing wire-bonding and assembly equipment means being on the fabrication floor.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Directing balance of plant, BOP) construction, generator installation, testing, commissioning or supervisory control and data acquisition
staying human
This work happens in the physical world: balance of plant, bop) construction, generator installation, testing, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Direct balance of plant (BOP) construction, generator installation, testing, commissioning, or supervisory control and data acquisition (SCADA) to ensure compliance with specifications.” (O*NET task statement)
How this row was scored
Exposure score: 6 out of 100 (2–10 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Directing construction, installation and commissioning happens on site with crews.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 1/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Installing, calibrating, operating or maintaining robots
staying human
This work happens in the physical world: robots, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Install, calibrate, operate, or maintain robots.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: the same decision, made over and over; work that happens in the physical world.
The rating behind it: Installing, calibrating and maintaining robots is physical work on the machine.
The five ratings: output a model can produce 1/4 · needs a body in a room 4/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Synthesizing, processing or characterize nanomaterials, using advanced tools or techniques
staying human
This work happens in the physical world: characterize nanomaterials, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Synthesize, process, or characterize nanomaterials, using advanced tools or techniques.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Making and characterizing nanomaterials is hands-on laboratory work.
The five ratings: output a model can produce 1/4 · needs a body in a room 4/4 · needs an accountable person 0/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.

What this job pays, and how many people do it

Median pay: $122,930a year, the middle of the range, so half earn more and half earn less.bls-oews, 2025 · May 2025 estimates (national_M2025_dl.xlsx)
How we know this
Source: bls-oews
Reference period: May 2025 estimates (national_M2025_dl.xlsx)
Rounding: Shown as published.
People doing this job: 154,070in the US, 2025.bls-oews · May 2025 estimates (national_M2025_dl.xlsx)

What is deliberately not here: a forecast of how many of these jobs exist in ten years. Where an official projection exists for a market we publish it with its vintage; where it does not, we leave the space empty rather than borrow the other country’s number.

Why this is shifting

The reason is boringly specific. Most of what is shifting here is reading one thing and writing another: energy savings strategies in, a record out. The rows above are exactly that shape: processing or interpreting signals or sensor data and creating plans for solar energy system development. What it cannot do is be there in the room, and that is still where characterize nanomaterials get done. Which is why this page talks about your tasks changing, not your job ending.

Your move

Over a pint: what I’d tell you if you were my friend

Your week is splitting in two, and which half fills it is the whole question. Processing or interpreting signals or sensor data is going; synthesizing, processing or characterize nanomaterials, using advanced tools or techniques is not.

So, given all that: 32% of this job's task weight sits in rows the software is already learning, 23% in rows that change shape rather than disappear, and 45% in rows it is nowhere near. That is the position, measured across 179 scored tasks. It is not a forecast about you.

The occupation is an average and you are not, so the first job below is finding out where your own week actually sits.

This week: one thing

Print the task ledger above and put your own hours next to each row. Roughly, in pen, in one sitting. Then look at which group your hours actually pile up in. Twenty minutes, this week.

What you end up holding: your own week, on one page, sorted into what is shifting and what is not
How long it takes: about twenty minutes

If printing it isn’t practical: Read the rows off this page and write the same list on the back of an envelope. Same twenty minutes, same page, and it works just as well said out loud to someone who knows the job. The point is your hours next to the rows, not the paper it is on.

Over the next 90 days

Volunteer toward the durable end, visibly. Over the next ninety days put your hand up for the work in the bottom group (synthesizing, processing or characterize nanomaterials, using advanced tools or techniques) and let people see you doing it. Not a new project: the same job, with your mix deliberately tilted. The point is that when the rota or the reorganisation comes, the version of you people picture is the one doing the part that stays.

Over the next 12 months

Claim a specialism at the durable end and let the other end go. Over a year, deliberately become the person who handles synthesizing, processing or characterize nanomaterials, using advanced tools or techniques, and deliberately stop being the first choice for the rows in the top group. That trade costs something, because the exposed work is often the comfortable work. Decide it on purpose rather than by drift. Before you pay for anything, use CareerOneStop - Find local training. It is free, it is the Labor Department's own service, and it is listed below with the rest of the free routes.

The roads out of here, and why I am not sending you down them

I looked at the obvious moves out of this job, and here is what I found.

I checked the 12 nearest US occupations to engineers, all other (nearest by the work that AI is not taking, not by job title), and none of them survived. The closest was industrial engineering technologists and technicians: only about 4% of its durable work is work you already do and it pays 46.2% less. Your own job splits about 32/68: that share of the list sits in the top exposure band and the rest does not. On this evidence the honest move is inside the job rather than out of it. Become the person who owns “conduct engineering site audits to collect structural, electrical, and related site information…”, and let the exposed end go.

How that was checked: this job was compared against all 830 US occupations in this release on their official task statements, and the 12 nearest were examined one by one. A move that turns on an industry, an employer or a qualification rather than on the work itself will not show up in a check like that. And this release carries no licence register, so anything you are weighing needs that looked up separately.

3 moves I checked and rejected

These are the obvious-looking jumps. They are here with their reasons rather than quietly dropped, because the ones that fail are worth knowing about. It is one less thing to turn over at night.

Industrial Engineering Technologists and Technicians
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already design or conduct tests of new nanotechnology products, processes, or systems, and their equivalent is to calibrate nanotechnology equipment. Across both published task lists that is about 4% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 4% of the durable side of that job. That is a different job, not a next step. It is a pay cut, in those words: $66,120 against your $122,930, 46.2% less (OEWS May 2025 (both)). Retraining to earn less is a decision, not advice.
Look at that job’s page anyway →
Electro-Mechanical and Mechatronics Technologists and Technicians
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already conduct research on robotic technology to create new robotic systems or system capabilities, and their equivalent is to build or assemble robotic devices or systems. Across both published task lists that is about 3% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 3% of the durable side of that job. That is a different job, not a next step. It is a pay cut, in those words: $73,900 against your $122,930, 39.9% less (OEWS May 2025 (both)). Retraining to earn less is a decision, not advice. And it is a narrow door: about 15,520 of those jobs against 154,070 of yours (OEWS May 2025), 10% as many seats.
Look at that job’s page anyway →
Industrial Engineers
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already supervise technologists, technicians, or other engineers, and their equivalent is to supervise technicians, technologists, analysts, administrative staff, or other engineers. Across both published task lists that is about 3% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 3% of the durable side of that job. That is a different job, not a next step. It is a pay cut, in those words: $102,440 against your $122,930, 16.7% less (OEWS May 2025 (both)). Retraining to earn less is a decision, not advice.
Look at that job’s page anyway →

What I’d stop worrying about

A friend tells you what not to spend fear on. This is that list.

The headline number you read somewhere
The big “X% of jobs” figures are about the whole economy, not about you. The number that describes your job is on this page: 32% of its task weight, across 179 scored tasks. Every row behind it is printed above with the source. Check ours; ignore theirs.
The whole-job doom story
Nothing on this page says this job ends. It says the mix inside it moves. Half the rows above are unchanged or changing shape, and the plan is about which half your week sits in. That is a very different problem, and a solvable one.
Panic-buying a course
Do the twenty-minute sorting exercise first. Most people who buy a course before they have done it buy the wrong one, and the free services listed below will tell you the same thing without charging for it.
The “obvious” next job everyone suggests
I checked the obvious moves and most of them did not survive. The reasons are printed with the routes above, including the pay and the gate. A move that fails on the numbers is worth knowing about so you can stop turning it over at night.

You are reading the United States figures

The United Kingdom splits this work across more than one official group, of which Engineering professionals n.e.c. is the closest. The pay and employment figures are not directly comparable, and we do not average them together.

Switch to the United Kingdom page →partial match

The other groups this work is counted across:

In UK official statistics this job is counted as Engineering professionals n.e.c., Quantity surveyors, Quality control and planning engineers and Engineering project managers and project engineers. Pay is shown separately for each of those groups (medians cannot be averaged together), while the task list and the scores on this page are for this group only.

Your route through this

Where to go next, and what it costs

Free, and complete

The moves above cost nothing. These are the real services that go with them: public, government-funded, and free at the point of use. Nothing on this page is behind an email address or a payment.

Anywhere in the US:
CareerOneStop - Find local training
Search what's running near you, from the Labor Department's own database, before anyone sells you a course.
Free to search; individual programs vary, and some are funded
Anywhere in the US:
American Job Center locator
An American Job Center will sit down with you for free. Find yours by ZIP code.
Free
- CareerOneStop local help hub
Anywhere in the US:
CareerOneStop - Licensed occupations finder
Check what your state actually requires before you pay for anything.
Free
Anywhere in the US:
CareerOneStop
Free

No Space for this job, but one for what is happening to it

Nothing Collab365 runs is built for engineers / all other, and we are not going to point you at the nearest one and call it a fit.

There is one that is not about a job title at all. The AI Authority is about being the person who directs these tools at work rather than the person they get compared to. That is worth saying here, because 32% of the work on this page is already inside what they can do.

Try The AI Authority free →

7 days free, no card needed. Explore up to 2 Spaces before you choose a plan: you pick a plan later, not now.

The AI Authority is a general community about working with AI, not a course for engineers / all other. You do not need it to act on anything here: the moves above cost nothing and stand on their own. The data on this page is the same either way.

Noted, and thank you. We’ll email you if a Space for engineers / all other launches. Nothing else.

That did not look like an email address, so nothing was saved. Have another go below.

We could not save that. The fault is ours, not yours, and your address was not stored. Please try again later.

No Space for engineers / all other yet. Should there be one?

Collab365 Spaces is built by the same people as this site. We find the problems that AI and automation are creating inside one kind of work, then solve them as short courses, briefings and Blueprints. Each Space is the community too, so the research and the people doing your job are in the same place.

What a Space actually is, in full

Collab365 launches new communities where the need is real. If one for engineers / all other existed, with researched problems, courses and people in the same boat, would you want in?

Your email

We use your email address for one thing: to tell you if a Space for engineers / all other launches. We never sell it, never use it for unrelated marketing, and every email has a one-click unsubscribe. Our privacy policy has the full version.

This unlocks nothing. Every figure, every row and every step on this page is already yours, whether you fill this in or not.

No deadline on any of this. The page will still be here, and the data is refreshed on a published schedule rather than when someone wants a headline.

Questions people ask about this job

Will AI replace Engineers, All Other?: Not as a job, but it is already doing parts of the work. Across the 179 official task statements scored for Engineers, All Other (United States, SOC 17-2199), 32% of the importance-weighted core work is made of tasks today's AI could already do most of. The overall exposure score is 46 out of 100 (range 40–52, band: partial). That is a statement about tasks, not about headcount: this measures what AI could do, not whether any employer adopts it, whether the law allows it, or whether doing the routine parts faster creates more demand for the human parts. Figures are from release 2026-q4.1.
Which tasks in “Engineers, All Other” can AI already do?: The highest-scoring tasks in release 2026-q4.1 are: “Maintain technical project files” (93/100, very high); “Develop customer documentation, such as performance specifications, training manuals, or operating instructions” (93/100, very high); “Make system device lists or event timing charts” (93/100, very high). Each score comes from five published 0–4 ratings turned into a number by a published formula, and each carries the model's one-sentence reason on the page.
Which tasks in “Engineers, All Other” stay human?: About 45% of this job's task weight sits in work that scores low for AI exposure. The lowest-scoring tasks in release 2026-q4.1 are: “Synthesize, process, or characterize nanomaterials, using advanced tools or techniques” (0/100, minimal); “Install, calibrate, operate, or maintain robots” (0/100, minimal); “Direct balance of plant (BOP) construction, generator installation, testing, commissioning, or supervisory control and data acquisition (SCADA) to ensure com…” (6/100, minimal). Low scores usually mean the task needs a body in a room, a legally accountable human, or trust built in real time. Those are the three things the scoring rubric treats as gates rather than obstacles.
What should someone working in “Engineers, All Other” do about AI?: Start from the ledger rather than the headline: 32% of this job's weighted core work is exposed, and roughly 45% is not. The practical move is to spend more of your week on the tasks that score low, the ones above, and to get fluent at directing AI through the tasks that score high, because those are the parts that change whether or not you are ready for them. This page does not predict your job, and nothing here is career advice tailored to you: the score describes the occupation, not the person.
How is the AI exposure score for Engineers, All Other calculated?: Each official task statement for the occupation is rated on five published 0–4 dimensions (output replicability, physical embodiment, licensed accountability, real-time human trust, and data availability) by claude-opus-5 using scoring prompt task_scoring_v1.0. The model never writes the score; a published formula turns the five ratings into a 0–100 number, so every score can be recomputed by hand. The occupation figure is the importance-weighted mean across 179 scored tasks. The prompt, the rubric, the formula and the full dataset are published at https://futureproof.collab365.com/method and https://futureproof.collab365.com/data/2026-q4.1 under CC BY 4.0.

Where these numbers come from

About the data on this page

One task scored differently between repeat runs, so its range on this page is wider. We would rather show the wobble than hide it.
The match between this job and its counterpart in the other country is partial, so the two pages count slightly different groups of people.
19 of this occupation's scored task statements carry a score that was measured under a different occupation's context, because the statement is shared between jobs and has only been scored once so far. Each one names the occupation it was measured under in the free routes below; none is presented as a measurement for this job.
Task clusters are not derived in this release, so the task-cluster field is empty and no Collab365 Space routing is attached to this occupation yet.

Task statements: onet-dbProcessing: catalogue-bridge → onet-im-rt-weighting → task-scoring → score-aggregation
Task weights: onet-db (im-rt-with-imputed)
Scores: Rubric task_scoring_v1.0, model claude-opus-5, computed 2026-08-04.
Pay and employment: bls-oews (May 2025 estimates (national_M2025_dl.xlsx))bls-oews (May 2025 estimates (national_M2025_dl.xlsx))

Figures on this page come from release 2026-q4.1, published 2026-08-05. Every release keeps its own permanent address, so a figure you cite in March is still there, unchanged, in November.

The plain-English wording on this page is assembled directly from the task statements and the published ratings, not written by hand for this occupation. That is why it is specific, and it is also why we say so.

The routes and free resources further up are today’s, not the release’s (last reviewed 2026-08-05). A route is an offer, not a historical fact, so it moves on its own clock.

How we score a job Download this release Look up another job

Using these figures?

Cite this

Everything on this site is published under CC BY 4.0. Quote it, chart it, sell something built on it. Just say where it came from, and cite the dated release rather than the site, so the figure you quote stays checkable.

Plain text

Collab365 (2026). Collab365 Futureproof: task-level AI exposure for US and UK occupations, release 2026-q4.1 (methodVersion 2.0.0, promptVersion task_scoring_v1.0). https://futureproof.collab365.com/data/2026-q4.1. Licensed CC BY 4.0. Built with O*NET data (USDOL/ETA, CC BY 4.0); ONS data (Open Government Licence v3.0); GAISI task framework (arXiv:2507.22748, MIT); BLS data (public domain).

BibTeX

@misc{collab365futureproof2026q41,
  title        = {Collab365 Futureproof: task-level AI exposure for US and UK occupations, release 2026-q4.1},
  author       = {{Collab365}},
  year         = {2026},
  url          = {https://futureproof.collab365.com/data/2026-q4.1},
  note         = {Release 2026-q4.1, methodVersion 2.0.0, promptVersion task_scoring_v1.0, CC BY 4.0}
}

Permalink

https://futureproof.collab365.com/data/2026-q4.1

Data as of release 2026-q4.1, published 2026-08-05. Releases never change after publication; when the figures move, a new dated release is published beside this one and this one stays exactly where it is.