Will AI replace Biological Scientists, All Other?

Not as a job, but it is already doing parts of the work. Across the 88 official task statements scored for Biological Scientists, All Other (United States, SOC 19-1029), 34% of the importance-weighted core work is made of tasks today's AI could already do most of. The overall exposure score is 47 out of 100 (range 42–53, 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 “Biological Scientists, All Other” can AI already do?

The highest-scoring tasks in release 2026-q4.1 are: “Program and use computers to store, process, and analyze data” (93/100, very high); “Prepare requests for proposals or statements of work” (93/100, very high); “Prepare summary statistics of information regarding human genomes” (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 “Biological Scientists, All Other” stay human?

About 47% 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: “Extract deoxyribonucleic acid (DNA) or perform diagnostic tests involving processes such as gel electrophoresis, Southern blot analysis, and polymerase chain…” (0/100, minimal); “Perform laboratory procedures following protocols including deoxyribonucleic acid (DNA) sequencing, cloning and extraction, ribonucleic acid (RNA) purificati…” (0/100, minimal); “Represent employer in a technical capacity at conferences” (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 “Biological Scientists, All Other” do about AI?

Start from the ledger rather than the headline: 34% of this job's weighted core work is exposed, and roughly 47% 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 Biological Scientists, 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 88 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

Biological Scientists, All Other

maintaining accurate laboratory records and data, providing scientific direction for project teams regarding the evaluation or handling of devices and developing and maintaining liaisons and effective working relations with groups and individuals. 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: evaluating genetic data by performing appropriate mathematical or statistical calculations and analyses. 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: 34%
changing shape: 19%
staying human: 47%

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

Whole-job exposure score 47 out of 100 (42–53 allowing for uncertainty): partial exposure, across 88 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 biological scientists, 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 2026-08-05. Read the full method.

That slice is not coming back; the core of the job, performing laboratory procedures following protocols, stays yours. The tools change, the responsibility doesn't.

Your week, as this page understands it

All biological scientists not listed separately. The job title says “biological scientists” or “all other”: officially one job, two names. The real job is the part underneath: performing laboratory procedures following protocols. 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 biological scientists, all other is not one task. It is 88 scored ones, and they are nowhere near equally exposed. The most durable of them, on this evidence, is performing laboratory procedures following protocols, and the ledger below shows exactly why.

Where the work sits, by task weight

shifting to AI: 34%
changing shape: 19%
staying human: 47%

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 2026-08-05. 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.

10 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

29 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.

Maintaining accurate laboratory records and data
This is reading one thing and writing another: accurate laboratory records in, a record out. That is the shape today's tools are built for.
importance 5 · Core
Source: “Maintain accurate laboratory records and data.” (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: Recording laboratory results accurately is routine record-keeping, though notes are often taken at the bench as work happens.
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.
Preparing results of experimental findings for presentation at professional conferences or in scientific journals
This is reading one thing and writing another: results of experimental findings in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Prepare results of experimental findings for presentation at professional conferences or in scientific journals.” (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: Turning findings into a paper or conference presentation is writing work that software drafts to a good starting standard.
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.
Searching scientific literature to select and modify methods and procedures most appropriate for genetic research goals
This is reading one thing and writing another: scientific literature in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Search scientific literature to select and modify methods and procedures most appropriate for genetic research goals.” (O*NET task statement)
How this row was scored
Exposure score: 83 out of 100 (76–90 allowing for uncertainty): very 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: Searching the literature and picking suitable methods is exactly the kind of retrieval and comparison AI handles 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.
Preparing or reviewing reports, manuscripts or meeting presentations
This is reading one thing and writing another: reports, manuscripts or meeting presentations in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Prepare or review reports, manuscripts, or meeting presentations.” (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: Drafting and reviewing manuscripts, reports and slide decks is writing work software now does to a good starting standard.
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.

Changing shape

15 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.

Writing grant applications to obtain funding
The software now makes the first pass at grant applications, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Write grant applications to obtain funding.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (54–62 allowing for uncertainty): partial exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Grant applications draft reasonably well, though a convincing one needs the researcher's own ideas and knowledge of the funder.
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.
Reviewing, approving or interpreting genetic laboratory results
The software now makes the first pass at genetic laboratory results, 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: “Review, approve, or interpret genetic laboratory results.” (O*NET task statement)
How this row was scored
Exposure score: 47 out of 100 (40–54 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: AI reads genetic results well, but approving and issuing them is a registered scientist's responsibility.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 3/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Communicating research results through conference presentations
The software now makes the first pass at research results, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Communicate research results through conference presentations, scientific publications, or project reports.” (O*NET task statement)
How this row was scored
Exposure score: 53 out of 100 (49–57 allowing for uncertainty): partial 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: Papers and reports draft well automatically, though presenting at a conference still means standing up in front of people.
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.

Staying human

44 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.

Designing molecular or cellular laboratory experiments
This work happens in the physical world: molecular, in a real place. Software cannot follow it there.
importance 5 · Core
Source: “Design molecular or cellular laboratory experiments, oversee their execution, and interpret results.” (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: Software can draft an experimental design and help read the results, but running the experiments is bench 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.
Supervising or directing the work of other geneticists
The value here is that a specific person handles the work of other geneticists and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Supervise or direct the work of other geneticists, biologists, technicians, or biometricians working on genetics research projects.” (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: Directing other scientists depends on judgment about people and a working relationship built face to face.
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.
Writing grants and papers or attending fundraising events to seek research funds
The ratings behind this row put grants well outside what today's tools can do on their own.
importance 4 · Core
Source: “Write grants and papers or attend fundraising events to seek research funds.” (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: Grant writing drafts reasonably well, but fundraising events and the researcher's own track record still carry the application.
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.

Show the other 78 tasks

Programing and using computers
shifting to AI
This is reading one thing and writing another: computers in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Program and use computers to store, process, and analyze 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: Storing, processing and analyzing data by writing code is among the things software now does most 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.
Preparing requests for proposals or statements of work
shifting to AI
This is reading one thing and writing another: requests in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Prepare requests for proposals or statements of work.” (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: Requests for proposals and statements of work follow standard structures that software fills in reliably from project details.
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.
Preparing summary statistics of information regarding human genomes
shifting to AI
This is reading one thing and writing another: summary statistics of information regarding human genomes in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Prepare summary statistics of information regarding human genomes.” (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: Producing summary statistics from genome data is calculation and reporting 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.
Compiling data for use in activities
shifting to AI
This is reading one thing and writing another: data in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Compile data for use in activities, such as gene expression profiling, genome annotation, or structural bioinformatics.” (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: Assembling data for gene expression or annotation work is scripted pipeline work that runs without a person watching.
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.
Manipulating publicly accessible, commercial or proprietary genomic, proteomic or posting-genomic databases
shifting to AI
This is reading one thing and writing another: accessible, commercial or proprietary genomic in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Manipulate publicly accessible, commercial, or proprietary genomic, proteomic, or post-genomic databases.” (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: Querying genomic and proteomic databases is structured search, though some of the databases are private to the employer.
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.
Evaluating genetic data by performing appropriate mathematical or statistical calculations and analyses
shifting to AI
This is reading one thing and writing another: genetic data in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Evaluate genetic data by performing appropriate mathematical or statistical calculations and analyses.” (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: Statistical analysis of genetic data follows established methods and is largely automated already.
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.
Testing new and updated bioinformatics tools and software
shifting to AI
This is reading one thing and writing another: new and updated bioinformatics tools and software in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Test new and updated bioinformatics tools and software.” (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: Testing software against expected behavior is routine, well-documented work that automated tools already do a great deal of.
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.
Developing new software applications or customize existing applications to meet specific scientific project needs
shifting to AI
This is reading one thing and writing another: new software applications in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Develop new software applications or customize existing applications to meet specific scientific project needs.” (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 and adapting scientific software is coding, and coding is one of the things current tools do best.
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.
Developing data models and databases
shifting to AI
This is reading one thing and writing another: data models in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Develop data models and databases.” (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: Designing data models and databases is well-documented technical work with abundant examples to draw on.
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.
Creating or modifying web-based bioinformatics tools
shifting to AI
This is reading one thing and writing another: web-based bioinformatics tools in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Create or modify web-based bioinformatics tools.” (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: Building and updating web-based research tools is ordinary software development with plenty of published examples.
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.
Designing, and applying bioinformaticsing algorithms
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 4 · Core
Source: “Design and apply bioinformatics algorithms including unsupervised and supervised machine learning, dynamic programming, or graphic algorithms.” (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: Bioinformatics algorithms and machine learning methods are heavily published, so writing and applying them is well within reach.
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.
Designing and maintaining genetics computer databases
shifting to AI
This is reading one thing and writing another: genetics computer databases in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Design and maintain genetics computer databases.” (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: Designing and maintaining genetics databases is standard data work with plenty of published practice to follow.
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.
Creating or using statistical models for the analysis of genetic data
shifting to AI
This is reading one thing and writing another: statistical models in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Create or use statistical models for the analysis of genetic data.” (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: Statistical modeling of genetic data is well-documented quantitative work with abundant published methods.
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.
Verifying all financial, physical and human resources assigned to research or development projects
shifting to AI
This is reading one thing and writing another: all financial, physical and human resources in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Verify all financial, physical, and human resources assigned to research or development projects are used as planned.” (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: Checking that project money, equipment and staff time were used as planned is straightforward records work.
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.
Designing databases, such as mutagenesis libraries
shifting to AI
This is reading one thing and writing another: databases in, a record out. That is the shape today's tools are built for.
importance 2 · Supplemental
Source: “Design databases, such as mutagenesis libraries.” (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: Designing a database for something like a mutagenesis library is standard data modeling work.
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.
Compiling and analyzing molecular or cellular experimental data and adjusting experimental designs
shifting to AI
This is reading one thing and writing another: molecular in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Compile and analyze molecular or cellular experimental data and adjust experimental designs as necessary.” (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: Compiling and analyzing experimental data, then flagging what to change, is analysis software handles well from the lab's own records.
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 large molecular datasets, such as raw microarray data
shifting to AI
This is reading one thing and writing another: large molecular datasets in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Analyze large molecular datasets, such as raw microarray data, genomic sequence data, or proteomics data, for clinical or basic research purposes.” (O*NET task statement)
How this row was scored
Exposure score: 72 out of 100 (68–76 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: Analyzing genomic and proteomic datasets is exactly what these pipelines were built for, with a scientist checking the conclusions.
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.
Improving user interfaces to bioinformatics software and databases
shifting to AI
This is reading one thing and writing another: user interfaces in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Improve user interfaces to bioinformatics software and databases.” (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: Interface improvements are standard software work, though knowing what actually annoys users takes some contact 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 1/4 · how much data exists 4/4.
Maintaining laboratory notebooks that record research methods
shifting to AI
This is reading one thing and writing another: laboratory notebooks in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Maintain laboratory notebooks that record research methods, procedures, and results.” (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: A lab notebook is a structured record of methods and results, which software captures well when entries are made at the bench.
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.
Preparing plans for management of renewable resources
shifting to AI
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 · Supplemental
Source: “Prepare plans for management of renewable resources.” (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: Resource management plans follow published frameworks and existing data, which software turns into solid drafts.
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 guidelines for procedures, the management of viruses
shifting to AI
This is reading one thing and writing another: guidelines in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Develop guidelines for procedures such as the management of viruses.” (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: Writing procedural guidelines from established practice is document work software drafts well, with a specialist approving the final version.
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 reports and proposals
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: “Review reports and proposals, such as those relating to land use classifications and recreational development, for accuracy, adequacy, or adherence to policies, regulations, or scientific standards.” (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: Reviewing reports against policies and scientific standards is document checking that software does thoroughly and quickly.
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 new systems and processes to improve operations
shifting to AI
This is reading one thing and writing another: new systems in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Recommend new systems and processes to improve operations.” (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: Recommending better systems and processes is analysis and writing that software drafts well from how the work currently runs.
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.
Providing statistical and computational tools for biologically based activities
shifting to AI
This is reading one thing and writing another: statistical in, a record out. That is the shape today's tools are built for.
importance 3 · Core
Source: “Provide statistical and computational tools for biologically based activities, such as genetic analysis, measurement of gene expression, or gene function determination.” (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: Building statistical and computational tools for colleagues is well-documented development work AI supports strongly.
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.
Studying basic principles of plant and animal life
shifting to AI
This is reading one thing and writing another: basic principles of plant in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Study basic principles of plant and animal life, such as origin, relationship, development, anatomy, and function.” (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: Basic biology is documented in enormous depth, so studying and synthesizing it is well suited to software.
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.
Creating novel computational approaches and analytical tools as required by research goals
changing shape
The software now makes the first pass at novel computational approaches, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Create novel computational approaches and analytical tools as required by research goals.” (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: Genuinely new computational methods still come from human insight, even though software can build and test the code quickly.
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.
Writing grant proposals to obtain funding for biological research
changing shape
The software now makes the first pass at grant proposals, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Write grant proposals to obtain funding for biological research.” (O*NET task statement)
How this row was scored
Exposure score: 58 out of 100 (54–62 allowing for uncertainty): partial exposure, high confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Grant proposals draft reasonably well, though a fundable one needs the researcher's own ideas and track record.
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.
Evaluating new technologies to enhance or complement current research
changing shape
The software now makes the first pass at new technologies, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Evaluate new technologies to enhance or complement current research.” (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: Comparing new instruments and methods against published specifications is research work software does thoroughly, with occasional hands-on trials.
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.
Preparing technical and researching reports
changing shape
The software now makes the first pass at technical and researching reports, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Prepare technical and research reports, such as environmental impact reports, and communicate the results to individuals in industry, government, or the general public.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (52–60 allowing for uncertainty): partial 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: Technical and impact reports are written outputs software drafts well, with the specialist checking figures before it goes out.
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 1/4 · how much data exists 3/4.
Keeping abreast of new biochemistries
changing shape
The software now makes the first pass at abreast of new biochemistries, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Keep abreast of new biochemistries, instrumentation, or software by reading scientific literature and attending professional conferences.” (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: Reading and summarizing new literature is something software does very quickly; conferences add a smaller in-person element.
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.
Collaborating with software developers in the development and modification of commercial bioinformatics software
changing shape
The software now makes the first pass at software developers, 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: “Collaborate with software developers in the development and modification of commercial bioinformatics software.” (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: Collaborating on commercial software is largely technical work at a screen, though it depends on working closely with a 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 2/4 · how much data exists 3/4.
Consulting with researchers to analyze problems
changing shape
The software now makes the first pass at researchers, 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: “Consult with researchers to analyze problems, recommend technology-based solutions, or determine computational strategies.” (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: Recommending a computational approach is well-documented advisory work, but researchers bring their own problem and need to be persuaded.
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.
Attending clinical and researching conferences and reading scientific literature to keep abreast of technological advances and current genetic research findings
changing shape
The software now makes the first pass at clinical, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Attend clinical and research conferences and read scientific literature to keep abreast of technological advances and current genetic research findings.” (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 published research is something software does exceptionally well; conferences add an in-person element.
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.
Developing pest management and controlling measures
changing shape
The software now makes the first pass at pest management, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 3 · Supplemental
Source: “Develop pest management and control measures, and conduct risk assessments related to pest exclusion, using scientific methods.” (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: reading one thing and writing another; the same decision, made over and over; mistakes that are cheap to catch.
The rating behind it: Pest control measures and risk assessments follow published methods and data, with occasional site inspection and specialist sign-off.
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 0/4 · how much data exists 3/4.
Researching environmental effects of present and potential uses of land and water areas
changing shape
The software now makes the first pass at environmental effects of present, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Supplemental
Source: “Research environmental effects of present and potential uses of land and water areas, determining methods of improving environmental conditions or such outputs as crop yields.” (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: Environmental effects research is largely analysis and modeling, though the underlying measurements come from real sites.
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.
Analyzing determinants responsible for specific inherited traits
changing shape
The software now makes the first pass at determinants responsible, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Analyze determinants responsible for specific inherited traits, and devise methods for altering traits or producing new traits.” (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: Working out which genes drive a trait is heavy analysis, but altering traits still requires real laboratory work.
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.
Conferring with information technology specialists to develop computer applications for genetic data analysis
changing shape
The software now makes the first pass at information technology specialists, 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: “Confer with information technology specialists to develop computer applications for genetic data analysis.” (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: Specifying analysis software with IT staff is desk work, though it depends on back-and-forth about what researchers actually need.
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.
Communicating test results to state and federal representatives and general public
staying human
The value here is that a specific person handles test results and stands behind it. That is earned, not computed.
importance 3 · Core
Source: “Communicate test results to state and federal representatives and general public.” (O*NET task statement)
How this row was scored
Exposure score: 39 out of 100 (32–46 allowing for uncertainty): low 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 results write up easily, but explaining them to officials and the public depends on being trusted in the room.
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 2/4 · how much data exists 3/4.
Instructing others in the selection and use of bioinformatics tools
staying human
The value here is that a specific person handles this work and stands behind it. That is earned, not computed.
importance 3 · Supplemental
Source: “Instruct others in the selection and use of bioinformatics tools.” (O*NET task statement)
How this row was scored
Exposure score: 39 out of 100 (32–46 allowing for uncertainty): low 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: Training material is easy to produce, though teaching people to use tools well involves working alongside them.
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 2/4 · how much data exists 3/4.
Developing protocols to improve existing genetic techniques or to incorporate new diagnostic procedures
staying human
The ratings behind this row put protocols well outside what today's tools can do on their own.
importance 3 · Core
Source: “Develop protocols to improve existing genetic techniques or to incorporate new diagnostic procedures.” (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: New protocols can be drafted from published methods, but proving they work needs bench validation before approval.
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.
Planning and administering biological research programs
staying human
The ratings behind this row put biological research programs well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Plan and administer biological research programs for government, research firms, medical industries, or manufacturing firms.” (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: Research programs can be planned on paper, but running one depends on internal knowledge and dealing with people.
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.
Coordinating molecular or cellular research activities with scientists specializing in other fields
staying human
The value here is that a specific person handles molecular and stands behind it. That is earned, not computed.
importance 3 · Core
Source: “Coordinate molecular or cellular research activities with scientists specializing in other fields.” (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 value is that a specific person does it.
The rating behind it: Coordinating across scientific fields happens through conversation and depends on relationships that a tool cannot stand in for.
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 2/4.
Conferring with vendors to evaluate new equipment or reagents or to discuss the customization of product lines to meet user requirements
staying human
The value here is that a specific person handles vendors and stands behind it. That is earned, not computed.
importance 3 · Supplemental
Source: “Confer with vendors to evaluate new equipment or reagents or to discuss the customization of product lines to meet user requirements.” (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 value is that a specific person does it.
The rating behind it: Vendor discussions are negotiations about custom requirements, where relationship and pricing knowledge matter more than published information.
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 2/4.
Planning curatorial programs for species collections that include acquisition
staying human
The ratings behind this row put curatorial programs well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Plan curatorial programs for species collections that include acquisition, distribution, maintenance, or regeneration.” (O*NET task statement)
How this row was scored
Exposure score: 32 out of 100 (25–39 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Curatorial planning follows documented practice, though what a specific collection needs depends on local knowledge and conditions.
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 2/4.
Participating in the development of endangered species breeding programs or species survival plans
staying human
The ratings behind this row put the development of endangered species breeding programs well outside what today's tools can do on their own.
importance 3 · Supplemental
Source: “Participate in the development of endangered species breeding programs or species survival plans.” (O*NET task statement)
How this row was scored
Exposure score: 32 out of 100 (25–39 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: mistakes that are cheap to catch.
The rating behind it: Breeding and survival plans draw on documented science, but the decisive details are local to each population and collection.
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 2/4.
Collaborating with biologists and other professionals to conduct appropriate genetic and biochemical analyses
staying human
The value here is that a specific person handles biologists and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Collaborate with biologists and other professionals to conduct appropriate genetic and biochemical analyses.” (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: Joint analyses across specialisms depend on colleagues sharing context and trusting each other's judgment.
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.
Collecting and analyzing biological data about relationships among and between organisms and their environment
staying human
This work happens in the physical world: biological data, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Collect and analyze biological data about relationships among and between organisms and their environment.” (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: The analysis half is well handled digitally, but collecting biological field data means going out and doing 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 0/4 · how much data exists 3/4.
Studying aquatic plants and animals and environmental conditions affecting them
staying human
This work happens in the physical world: aquatic plants, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Study aquatic plants and animals and environmental conditions affecting them, such as radioactivity or pollution.” (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: Studying aquatic life and pollution needs sampling in the water as well as analysis at a desk.
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.
Identifying, classifying and studying structure, behavior, ecology, physiology, nutrition, culture and distribution of plant and animal species
staying human
This work happens in the physical world: structure, behavior, ecology, physiology, nutrition, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Identify, classify, and study structure, behavior, ecology, physiology, nutrition, culture, and distribution of plant and animal species.” (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: Identification and classification draw on huge documented reference material, but studying species means fieldwork and specimens.
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.
Studying and managing wild animal populations
staying human
This work happens in the physical world: wild animal populations, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Study and manage wild animal populations.” (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: Wildlife population work combines desk analysis with counts, tagging and field observation that has to be done outdoors.
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 new supplies and equipment to ensure operability in specific laboratory settings
staying human
This work happens in the physical world: new supplies, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Evaluate new supplies and equipment to ensure operability in specific laboratory settings.” (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: Specifications are easy to compare, but checking equipment works in your own lab means trying 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 0/4 · how much data exists 3/4.
Instructing undergraduate and graduate students within the areas of cellular or molecular biology
staying human
This work happens in the physical world: undergraduate, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Instruct undergraduate and graduate students within the areas of cellular or molecular biology.” (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; work that happens in the physical world; the value is that a specific person does it.
The rating behind it: The subject matter is thoroughly documented, but teaching students well depends on being in the room and responding to them.
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 2/4 · how much data exists 4/4.
Conducting research on cell organization and function
staying human
This work happens in the physical world: research, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Conduct research on cell organization and function, including mechanisms of gene expression, cellular bioinformatics, cell signaling, or cell differentiation.” (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: Cell biology research mixes bench experiments with analysis, and only the analysis half can be handled at a screen.
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.
Conducting applied research aimed at improvements in areas
staying human
This work happens in the physical world: applied research, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Conduct applied research aimed at improvements in areas such as disease testing, crop quality, pharmaceuticals, and the harnessing of microbes to recycle waste.” (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: Applied research on disease tests or crop quality needs real experiments as well as analysis and reading.
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.
Instructing medical students, graduate students or others in methods or procedures for diagnosis and management of genetic disorders
staying human
This work happens in the physical world: medical students, graduate students, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Instruct medical students, graduate students, or others in methods or procedures for diagnosis and management of genetic disorders.” (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; work that happens in the physical world; the value is that a specific person does it.
The rating behind it: The material is thoroughly documented, but teaching diagnosis and management to medical students happens largely in person.
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 2/4 · how much data exists 4/4.
Planning or conducting basic genomic and biological research
staying human
This work happens in the physical world: basic genomic, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Plan or conduct basic genomic and biological research related to areas such as regulation of gene expression, protein interactions, metabolic networks, and nucleic acid or protein complexes.” (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: Genomic research combines literature and analysis with real laboratory work that has to be done by hand.
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.
Conferring with departments, such as marketing, business development or operations, to coordinate product development or improvement
staying human
The value here is that a specific person handles departments and stands behind it. That is earned, not computed.
importance 3 · Supplemental
Source: “Confer with departments, such as marketing, business development, or operations, to coordinate product development or improvement.” (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 value is that a specific person does it.
The rating behind it: Coordinating with marketing or operations happens in meetings where the useful information is internal and the relationships matter.
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 2/4.
Directing the work of technicians and information technology staff applying bioinformatics tools or applications in areas
staying human
The value here is that a specific person handles the work of technicians and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Direct the work of technicians and information technology staff applying bioinformatics tools or applications in areas such as proteomics, transcriptomics, metabolomics, or clinical bioinformatics.” (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 value is that a specific person does it.
The rating behind it: Directing technicians depends on knowing the people and the project, and much of it happens face to face.
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 2/4.
Directing, coordinating, organizing or prioritize biological laboratory activities
staying human
The value here is that a specific person handles prioritize biological laboratory activities and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Direct, coordinate, organize, or prioritize biological laboratory activities.” (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 value is that a specific person does it.
The rating behind it: Running a lab's day-to-day priorities depends on knowing the people, the equipment and what is actually urgent.
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 2/4.
Developing methods and apparatus for securing representative plant
staying human
This work happens in the physical world: methods, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Develop methods and apparatus for securing representative plant, animal, aquatic, or soil samples.” (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: Designing sampling apparatus involves building and testing real equipment in the conditions where it will be used.
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.
Designing sampling plans or coordinating the field collection of samples
staying human
This work happens in the physical world: plans, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Design sampling plans or coordinate the field collection of samples such as tissue specimens.” (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: Sampling plans can be designed on paper, but collecting tissue specimens in the field means being there.
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.
Providing scientific direction for project teams regarding the evaluation or handling of devices
staying human
The value here is that a specific person handles scientific direction and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Provide scientific direction for project teams regarding the evaluation or handling of devices, drugs, or cells for in vitro and in vivo disease models.” (O*NET task statement)
How this row was scored
Exposure score: 23 out of 100 (16–30 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the value is that a specific person does it.
The rating behind it: Steering a project team on how to handle cells or drugs rests on experience and on being trusted by the team.
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 2/4.
Conducting family medical studies to evaluate the genetic basis for traits or diseases
staying human
The value here is that a specific person handles medical studies and stands behind it. That is earned, not computed.
importance 3 · Supplemental
Source: “Conduct family medical studies to evaluate the genetic basis for traits or diseases.” (O*NET task statement)
How this row was scored
Exposure score: 23 out of 100 (16–30 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the value is that a specific person does it.
The rating behind it: Family medical studies rely on relatives sharing private history in conversation, which is the hard part of the work.
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 2/4.
Maintaining laboratory safety programs and training personnel in laboratory safety techniques
staying human
This work happens in the physical world: laboratory safety programs, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Maintain laboratory safety programs and train personnel in laboratory safety techniques.” (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: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Safety documents draft easily, but running a safety program and training people means being present in the laboratory.
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 1/4 · how much data exists 3/4.
Teaching or supervising students and performing research at universities and colleges
staying human
This work happens in the physical world: students, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Teach or supervise students and perform research at universities and colleges.” (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: University teaching and supervision happen in person and depend on the relationship between student and researcher.
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.
Participating in all levels of bioproduct development
staying human
This work happens in the physical world: all levels of bioproduct development, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Participate in all levels of bioproduct development, including proposing new products, performing market analyses, designing and performing experiments, and collaborating with operations and quality control teams during product launches.” (O*NET task statement)
How this row was scored
Exposure score: 18 out of 100 (11–25 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world; the value is that a specific person does it.
The rating behind it: Product development spans market analysis, bench experiments and launch meetings, so only part of it sits at a screen.
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 2/4.
Studying reactions of plants, animals and marine species to parasites
staying human
This work happens in the physical world: reactions of plants, animals and marine species, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Study reactions of plants, animals, and marine species to parasites.” (O*NET task statement)
How this row was scored
Exposure score: 14 out of 100 (7–21 allowing for uncertainty): minimal 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: Studying how species react to parasites means observing and testing living things directly.
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.
Developing and maintaining liaisons and effective working relations with groups and individuals
staying human
The value here is that a specific person handles liaisons and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Develop and maintain liaisons and effective working relations with groups and individuals, agencies, and the public to encourage cooperative management strategies or to develop information and interpret findings.” (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: Building working relationships with agencies and the public is the whole point of the task and cannot be handed over.
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.
Supervising biological technicians and technologists and other scientists
staying human
The value here is that a specific person handles biological technicians and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Supervise biological technicians and technologists and other scientists.” (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 technicians and scientists is about developing people and depends on a relationship built over time.
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.
Supervising technical personnel and postdoctoral research fellows
staying human
The value here is that a specific person handles technical personnel and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Supervise technical personnel and postdoctoral research fellows.” (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 is about developing people, which depends on a working relationship built over time.
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.
Measuring salinity, acidity, light, oxygen content and other physical conditions of water to determine their relationship to aquatic life
staying human
This work happens in the physical world: salinity, acidity, light, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Measure salinity, acidity, light, oxygen content, and other physical conditions of water to determine their relationship to aquatic life.” (O*NET task statement)
How this row was scored
Exposure score: 10 out of 100 (6–14 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: Water measurements are taken with instruments at the water's edge, even though interpreting them is straightforward.
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 3/4.
Monitoring or operating specialized equipment
staying human
This work happens in the physical world: specialized equipment, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Monitor or operate specialized equipment, such as gas chromatographs and high pressure liquid chromatographs, electrophoresis units, thermocyclers, fluorescence activated cell sorters, and phosphorimagers.” (O*NET task statement)
How this row was scored
Exposure score: 10 out of 100 (3–17 allowing for uncertainty): minimal exposure, medium 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: Instruments like chromatographs and cell sorters have to be loaded, run and watched by someone in the lab.
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 3/4.
Developing assays that monitor cell characteristics
staying human
This work happens in the physical world: assays, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Develop assays that monitor cell characteristics.” (O*NET task statement)
How this row was scored
Exposure score: 10 out of 100 (3–17 allowing for uncertainty): minimal exposure, medium 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: Developing an assay means building and testing it physically with real cells.
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 3/4.
Verifying that cytogenetic, molecular genetic and related equipment and instrumentation is maintained in working condition to ensure accuracy and quality of experimental results
staying human
This work happens in the physical world: cytogenetic, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Verify that cytogenetic, molecular genetic, and related equipment and instrumentation is maintained in working condition to ensure accuracy and quality of experimental results.” (O*NET task statement)
How this row was scored
Exposure score: 10 out of 100 (3–17 allowing for uncertainty): minimal exposure, medium 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: Confirming that genetic instruments are working properly means checking and calibrating the machines themselves.
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 3/4.
Evaluating, diagnosing or treating genetic diseases
staying human
This work happens in the physical world: genetic diseases, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Evaluate, diagnose, or treat genetic diseases.” (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: the same decision, made over and over; work that happens in the physical world; someone qualified has to answer for it; the value is that a specific person does it.
The rating behind it: Diagnosing and treating genetic disease is legally a licensed clinician's act, done with the patient and their family.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 4/4 · needs to be trusted in the moment 3/4 · how much data exists 3/4.
Representing employer in a technical capacity at conferences
staying human
This work happens in the physical world: employer, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Represent employer in a technical capacity at conferences.” (O*NET task statement)
How this row was scored
Exposure score: 6 out of 100 (0–13 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world; the value is that a specific person does it.
The rating behind it: Representing an employer at a conference means a named person being there to speak and answer for the organization.
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 2/4 · how much data exists 2/4.
Performing laboratory procedures following protocols
staying human
This work happens in the physical world: laboratory procedures following protocols, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Perform laboratory procedures following protocols including deoxyribonucleic acid (DNA) sequencing, cloning and extraction, ribonucleic acid (RNA) purification, or gel electrophoresis.” (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: Sequencing, cloning and gel work are done by hand at the bench with real samples and reagents.
The five ratings: output a model can produce 0/4 · 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.
Extracting deoxyribonucleic acid, DNA) or performing diagnostic tests involving processes, gel electrophoresis
staying human
This work happens in the physical world: deoxyribonucleic acid, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Extract deoxyribonucleic acid (DNA) or perform diagnostic tests involving processes such as gel electrophoresis, Southern blot analysis, and polymerase chain reaction analysis.” (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: DNA extraction, electrophoresis and PCR are physical bench procedures performed on real samples.
The five ratings: output a model can produce 0/4 · 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.

What this job pays, and how many people do it

Median pay: $98,920a 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: 55,850in 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: genetic data in, a record out. The rows above are exactly that shape: evaluating genetic data by performing appropriate mathematical or statistical calculations and analyses and maintaining accurate laboratory records and data. What it cannot do is be there in the room, and that is still where laboratory procedures following protocols 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. Evaluating genetic data by performing appropriate mathematical or statistical calculations and analyses is going; performing laboratory procedures following protocols is not.

So, given all that: 34% of this job's task weight sits in rows the software is already learning, 19% in rows that change shape rather than disappear, and 47% in rows it is nowhere near. That is the position, measured across 88 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 (performing laboratory procedures following protocols) 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 performing laboratory procedures following protocols, 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 biological scientists, all other (nearest by the work that AI is not taking, not by job title), and none of them survived. The closest was genetic counselors: only about 2% of its durable work is work you already do and there are far fewer of those jobs than of yours. Your own job splits about 34/66: 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 “design molecular or cellular laboratory experiments, oversee their execution, and interpret results”, 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.

Genetic Counselors
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already review, approve, or interpret genetic laboratory results, and their equivalent is to interpret laboratory results and communicate findings to patients or physicians. Across both published task lists that is about 2% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 2% of the durable side of that job. That is a different job, not a next step. And it is a narrow door: about 3,740 of those jobs against 55,850 of yours (OEWS May 2025), 7% as many seats.
Look at that job’s page anyway →
Physicists
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already prepare results of experimental findings for presentation at professional conferences or in scientific…, and their equivalent is to report experimental results by writing papers for scientific journals or by presenting information…. Across both published task lists that is about 2% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 2% of the durable side of that job. That is a different job, not a next step.
Look at that job’s page anyway →
Anthropology and Archeology Teachers, Postsecondary
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already prepare results of experimental findings for presentation at professional conferences or in scientific…, and their equivalent is to conduct research in a particular field of knowledge and present findings in professional…. Across both published task lists that is about 2% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 2% of the durable side of that job. That is a different job, not a next step. And it is a narrow door: about 5,240 of those jobs against 55,850 of yours (OEWS May 2025), 9% as many seats.
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: 34% of its task weight, across 88 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 Biochemists and biomedical scientists 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 Biochemists and biomedical scientists, Natural and social science professionals n.e.c., Other researchers, unspecified discipline and Biological scientists. 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 biological scientists / 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 34% 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 biological scientists / 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 biological scientists / 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 biological scientists / 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 biological scientists / 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 biological scientists / 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 Biological Scientists, All Other?: Not as a job, but it is already doing parts of the work. Across the 88 official task statements scored for Biological Scientists, All Other (United States, SOC 19-1029), 34% of the importance-weighted core work is made of tasks today's AI could already do most of. The overall exposure score is 47 out of 100 (range 42–53, 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 “Biological Scientists, All Other” can AI already do?: The highest-scoring tasks in release 2026-q4.1 are: “Program and use computers to store, process, and analyze data” (93/100, very high); “Prepare requests for proposals or statements of work” (93/100, very high); “Prepare summary statistics of information regarding human genomes” (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 “Biological Scientists, All Other” stay human?: About 47% 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: “Extract deoxyribonucleic acid (DNA) or perform diagnostic tests involving processes such as gel electrophoresis, Southern blot analysis, and polymerase chain…” (0/100, minimal); “Perform laboratory procedures following protocols including deoxyribonucleic acid (DNA) sequencing, cloning and extraction, ribonucleic acid (RNA) purificati…” (0/100, minimal); “Represent employer in a technical capacity at conferences” (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 “Biological Scientists, All Other” do about AI?: Start from the ledger rather than the headline: 34% of this job's weighted core work is exposed, and roughly 47% 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 Biological Scientists, 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 88 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

Worth knowing about these figures

The match between this job and its counterpart in the other country is partial, so the two pages count slightly different groups of people.
10 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)
Scores: Rubric task_scoring_v1.0, model claude-opus-5, computed 2026-08-05.
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.