Will AI replace Detectives and Criminal Investigators?

Not as a job, but it is already doing parts of the work. Across the 67 official task statements scored for Detectives and Criminal Investigators (United States, SOC 33-3021), 29% of the importance-weighted core work is made of tasks today's AI could already do most of. The overall exposure score is 32 out of 100 (range 28–38, band: low). 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 “Detectives and Criminal Investigators” can AI already do?

The highest-scoring tasks in release 2026-q4.1 are: “Evaluate records of communications, such as telephone calls, to plot activity and determine the size and location of criminal groups and members” (81/100, very high); “Gather, analyze, correlate, or evaluate information from a variety of resources, such as law enforcement databases” (81/100, very high); “Link or chart suspects to criminal organizations or events to determine activities and interrelationships” (81/100, 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 “Detectives and Criminal Investigators” stay human?

About 62% 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: “Take fingerprints” (0/100, minimal); “Maintain surveillance of establishments to obtain identifying information on suspects” (0/100, minimal); “Search for and collect evidence, such as fingerprints, using investigative equipment” (0/100, 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 “Detectives and Criminal Investigators” do about AI?

Start from the ledger rather than the headline: 29% of this job's weighted core work is exposed, and roughly 62% 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 Detectives and Criminal Investigators 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 67 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

Detectives and Criminal Investigators

checking victims for signs of life, searching for and collecting evidence and preparing and serving search and arrest warrants. 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: gathering, analyzing, correlate or evaluating information from a variety of resources, such as law enforcement databases. 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: 29%
changing shape: 9%
staying human: 62%

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

Whole-job exposure score 32 out of 100 (28–38 allowing for uncertainty): low exposure, across 67 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 detectives and criminal investigators is rated on five dimensions: can a model produce the output, does the work need a body in a room, does it need a legally accountable person, does it depend on a person being trusted in the moment, and how much data exists. A published formula turns those five ratings into the score; the model never writes the number.

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

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

That slice is not coming back; the core of the job, testifying in court and presenting evidence, stays yours. New tools, same person answering for it.

Your week, as this page understands it

Conduct investigations related to suspected violations of federal, state, or local laws to prevent or solve crimes. The job title says “detectives” or “criminal investigators”: officially one job, two names. The real job is the part underneath: testifying in court and presenting evidence. 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 detectives and criminal investigators is not one task. It is 67 scored ones, and they are nowhere near equally exposed. The most durable of them, on this evidence, is testifying in court and presenting evidence, and the ledger below shows exactly why.

Where the work sits, by task weight

shifting to AI: 29%
changing shape: 9%
staying human: 62%

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

How we know this

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

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

Your job, task by task

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

One task scored differently between repeat runs, so its range on this page is wider. We would rather show the wobble than hide it.
3 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

18 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.

Recording progress of investigation
This is reading one thing and writing another: progress of investigation in, a record out. That is the shape today's tools are built for.
importance 5 · Core
Source: “Record progress of investigation, maintain informational files on suspects, and submit reports to commanding officer or magistrate to authorize warrants.” (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: Case logs, suspect files and warrant reports are structured writing that software drafts well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Gathering, analyzing, correlate or evaluating information from a variety of resources, such as law enforcement databases
This is reading one thing and writing another: correlate in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Gather, analyze, correlate, or evaluate information from a variety of resources, such as law enforcement databases.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Pulling together and weighing information from many databases is core software territory.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Preparing reports that detail investigation findings
This is reading one thing and writing another: reports in, a record out. That is the shape today's tools are built for.
importance 5 · Core
Source: “Prepare reports that detail investigation findings.” (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: Turning investigation findings into a written report is well-suited to drafting software.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Validating known intelligence with data from other sources
This is reading one thing and writing another: known intelligence in, a record out. That is the shape today's tools are built for.
importance 5 · Core
Source: “Validate known intelligence with data from other sources.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Cross-checking one source against others is a search-and-compare job software does well.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.

Changing shape

6 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.

Preparing charges or responses
The software now makes the first pass at charges, 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 5 · Core
Source: “Prepare charges or responses to charges, or information for court cases, according to formalized procedures.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; someone qualified has to answer for it.
The rating behind it: Charge papers follow set formats, but a sworn officer or prosecutor must stand behind them.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Collaborating with other offices and agencies to exchange information and coordinating activities
The software now makes the first pass at other offices, 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: “Collaborate with other offices and agencies to exchange information and coordinate activities.” (O*NET task statement)
How this row was scored
Exposure score: 41 out of 100 (34–48 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the value is that a specific person does it.
The rating behind it: Sharing information between agencies is largely written, though trust between contacts still matters.
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 2/4 · how much data exists 2/4.
Notifying or requesting notification
The software now makes the first pass at notification, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Notify, or request notification of, medical examiner or district attorney representative.” (O*NET task statement)
How this row was scored
Exposure score: 52 out of 100 (45–59 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: Notifying a medical examiner or prosecutor is routine, rule-driven communication.
The five ratings: output a model can produce 4/4 · needs a body in a room 1/4 · needs an accountable person 1/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.

Staying human

43 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.

Obtaining facts or statements from complainants
This work happens in the physical world: facts, in a real place. Software cannot follow it there.
importance 5 · Core
Source: “Obtain facts or statements from complainants, witnesses, and accused persons and record interviews, using recording device.” (O*NET task statement)
How this row was scored
Exposure score: 5 out of 100 (0–12 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: 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: Recordings transcribe easily, but getting people to talk depends on a sworn officer being present.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 2/4 · needs to be trusted in the moment 3/4 · how much data exists 2/4.
Testifying in court and presenting evidence
This work happens in the physical world: court, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Testify in court and present evidence.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it; the value is that a specific person does it.
The rating behind it: Giving evidence in court is a sworn person appearing in the courtroom.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 4/4 · needs to be trusted in the moment 2/4 · how much data exists 2/4.
Obtaining summary of incident from officer in charge at crime scene
This work happens in the physical world: summary of incident, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Obtain summary of incident from officer in charge at crime scene, taking care to avoid disturbing evidence.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; the value is that a specific person does it.
The rating behind it: Taking a handover at a crime scene without disturbing it means being at that scene.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 2/4 · how much data exists 2/4.

Show the other 57 tasks

Evaluating records of communications, such as telephone calls
shifting to AI
This is reading one thing and writing another: records of communications in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Evaluate records of communications, such as telephone calls, to plot activity and determine the size and location of criminal groups and members.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Mapping call records to work out group size and location is straightforward data analysis.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Linking or charting suspects to criminal organizations or events to determine activities and interrelationships
shifting to AI
This is reading one thing and writing another: suspects in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Link or chart suspects to criminal organizations or events to determine activities and interrelationships.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Charting who links to whom is a mapping task software handles very well.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Preparing comprehensive written reports, presentations, maps or charts, based on research, collection and analysis of intelligence data
shifting to AI
This is reading one thing and writing another: comprehensive written reports, presentations, maps or charts in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Prepare comprehensive written reports, presentations, maps, or charts, based on research, collection, and analysis of intelligence data.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Turning analyzed data into reports, maps and charts is well-covered automated output.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Studying the assets of criminal suspects to determine the flow of money from or to targeted groups
shifting to AI
This is reading one thing and writing another: the assets of criminal suspects in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Study the assets of criminal suspects to determine the flow of money from or to targeted groups.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Tracing money between accounts and groups is transaction analysis software is very good at.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Analyzing intelligence data to identify patterns and trends in criminal activity
shifting to AI
This is reading one thing and writing another: intelligence data in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Analyze intelligence data to identify patterns and trends in criminal activity.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Finding patterns and trends in crime data is exactly what analysis software is built for.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Examining records to locate links in chains of evidence or information
shifting to AI
This is reading one thing and writing another: records in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Examine records to locate links in chains of evidence or information.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Finding links across records is exactly the kind of pattern search software excels at.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Identifying gaps in information
shifting to AI
This is reading one thing and writing another: gaps in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Identify gaps in information.” (O*NET task statement)
How this row was scored
Exposure score: 81 out of 100 (74–88 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: Spotting what is missing from a case file is a comparison job software does thoroughly.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Designing, using or maintaining databases and software applications
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 4 · Core
Source: “Design, use, or maintain databases and software applications, such as geographic information systems (GIS) mapping and artificial intelligence tools.” (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: Building and maintaining databases and mapping tools is documented software 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.
Examining records and governmental agency files to find identifying data about suspects
shifting to AI
This is reading one thing and writing another: records in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Examine records and governmental agency files to find identifying data about suspects.” (O*NET task statement)
How this row was scored
Exposure score: 69 out of 100 (62–76 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; someone qualified has to answer for it.
The rating behind it: Searching agency records for identifying details is database work, though access is restricted to cleared staff.
The five ratings: output a model can produce 4/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Studying activities relating
shifting to AI
This is reading one thing and writing another: activities in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Study activities relating to narcotics, money laundering, gangs, auto theft rings, terrorism, or other national security threats.” (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: Studying how crime networks operate is research and reading across large document sets.
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.
Studying communication code languages or foreign languages to translate intelligence
shifting to AI
This is reading one thing and writing another: communication code languages in, a record out. That is the shape today's tools are built for.
importance 3 · Supplemental
Source: “Study communication code languages or foreign languages to translate intelligence.” (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: Machine translation is strong for common languages, though coded material still needs specialists.
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.
Maintaining records of evidence and writing and reviewing reports
shifting to AI
This is reading one thing and writing another: records of evidence in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Maintain records of evidence and write and review reports.” (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: Maintaining evidence records and writing reports is structured documentation work.
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.
Analyzing completed police reports to determine what additional information and investigative work
shifting to AI
This is reading one thing and writing another: completed police reports in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Analyze completed police reports to determine what additional information and investigative work is needed.” (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: Reading reports to work out what is still missing is analysis software does well.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Identifying case issues and evidence
shifting to AI
This is reading one thing and writing another: case issues in, a record out. That is the shape today's tools are built for.
importance 4 · Core
Source: “Identify case issues and evidence needed, based on analysis of charges, complaints, or allegations of law violations.” (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: Working out what a case needs from the charges and complaints is document analysis.
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.
Predicting future gang, organized crime or terrorist activity, using analyses of intelligence data
changing shape
The software now makes the first pass at future gang, organized crime or terrorist activity, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Predict future gang, organized crime, or terrorist activity, using analyses of intelligence data.” (O*NET task statement)
How this row was scored
Exposure score: 59 out of 100 (52–66 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; mistakes that are cheap to catch.
The rating behind it: Forecasting from data is a strength, but the decisive local intelligence is private and patchy.
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 2/4.
Establishing criminal profiles to aid in connecting criminal organizations with their members
changing shape
The software now makes the first pass at criminal profiles, but the deciding part still needs a person. So the job becomes checking and deciding rather than producing.
importance 4 · Core
Source: “Establish criminal profiles to aid in connecting criminal organizations with their members.” (O*NET task statement)
How this row was scored
Exposure score: 59 out of 100 (52–66 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; mistakes that are cheap to catch.
The rating behind it: Profiles can be drafted from data, but much of the useful detail is unwritten local knowledge.
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 2/4.
Identifying, comparing, classifying and filing fingerprints, using systems, Automated Fingerprint Identification System
changing shape
The software now makes the first pass at fingerprints, 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 · Supplemental
Source: “Identify, compare, classify, and file fingerprints, using systems such as Automated Fingerprint Identification System (AFIS) or the Henry Classification System.” (O*NET task statement)
How this row was scored
Exposure score: 56 out of 100 (49–63 allowing for uncertainty): partial exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the same decision, made over and over; someone qualified has to answer for it.
The rating behind it: Automated matching does most of the comparison, though a certified examiner confirms an identification.
The five ratings: output a model can produce 3/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Conducting presentations of analytic findings
staying human
The value here is that a specific person handles presentations of analytic findings and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Conduct presentations of analytic findings.” (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 slides write themselves, but standing up and taking questions is still a person’s job.
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.
Determining scope, timing and direction of investigations
staying human
The rules require a named, qualified person to answer for scope, timing and direction of investigations, and that person cannot be a piece of software.
importance 4 · Core
Source: “Determine scope, timing, and direction of investigations.” (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: someone qualified has to answer for it.
The rating behind it: Deciding how far and how fast an investigation goes is judgment a sworn investigator owns.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Providing information to lab personnel concerning the source of an item of evidence and tests
staying human
The rules require a named, qualified person to answer for information, and that person cannot be a piece of software.
importance 4 · Core
Source: “Provide information to lab personnel concerning the source of an item of evidence and tests to be performed.” (O*NET task statement)
How this row was scored
Exposure score: 36 out of 100 (29–43 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; someone qualified has to answer for it.
The rating behind it: Telling the laboratory where an item came from and what tests are wanted is short, structured form work.
The five ratings: output a model can produce 3/4 · needs a body in a room 1/4 · needs an accountable person 2/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Developing defense plans or tactics
staying human
The rules require a named, qualified person to answer for defense plans, and that person cannot be a piece of software.
importance 4 · Supplemental
Source: “Develop defense plans or tactics, using intelligence and other information.” (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: someone qualified has to answer for it.
The rating behind it: Tactical plans lean on unwritten operational knowledge that a responsible officer must own.
The five ratings: output a model can produce 2/4 · needs a body in a room 0/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Serving as technical advisor and coordinating with other law enforcement workers or legal personnel to exchange information on crime scene collection activities
staying human
The value here is that a specific person handles technical advisor and stands behind it. That is earned, not computed.
importance 4 · Core
Source: “Serve as technical advisor and coordinate with other law enforcement workers or legal personnel to exchange information on crime scene collection activities.” (O*NET task statement)
How this row was scored
Exposure score: 31 out of 100 (24–38 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: reading one thing and writing another; the value is that a specific person does it.
The rating behind it: Advising colleagues on scene-collection practice is largely explanation, though relationships shape how it lands.
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 2/4 · how much data exists 2/4.
Gathering and evaluating information, using tools, aerial photographs, radar equipment or sensitive radio equipment
staying human
This work happens in the physical world: information, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Gather and evaluate information, using tools such as aerial photographs, radar equipment, or sensitive radio equipment.” (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: mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Image analysis is a strength, but operating radar and radio equipment means being with it.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Submitting evidence to supervisors, crime labs or court officials for legal proceedings
staying human
This work happens in the physical world: evidence, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Submit evidence to supervisors, crime labs, or court officials for legal proceedings.” (O*NET task statement)
How this row was scored
Exposure score: 28 out of 100 (21–35 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: The paperwork is routine, but the chain of custody means a named person hands the items over.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Preparing plans to intercept foreign communications transmissions
staying human
The rules require a named, qualified person to answer for plans, and that person cannot be a piece of software.
importance 3 · Supplemental
Source: “Prepare plans to intercept foreign communications transmissions.” (O*NET task statement)
How this row was scored
Exposure score: 27 out of 100 (20–34 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: someone qualified has to answer for it.
The rating behind it: Interception planning is restricted to authorised personnel and rests on unpublished technical detail.
The five ratings: output a model can produce 2/4 · 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 1/4.
Coordinating or conducting instructional classes or in-services
staying human
This work happens in the physical world: instructional classes, in a real place. Software cannot follow it there.
importance 3 · Core
Source: “Coordinate or conduct instructional classes or in-services, such as citizen police academy classes and crime scene training for other officers.” (O*NET task statement)
How this row was scored
Exposure score: 26 out of 100 (19–33 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; work that happens in the physical world; the value is that a specific person does it.
The rating behind it: Course material is easy to produce, but classes are delivered in a room with people.
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 3/4.
Creating sketches and diagrams, by hand or computer software, to depict crime scenes
staying human
This work happens in the physical world: sketches, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Create sketches and diagrams, by hand or computer software, to depict crime scenes.” (O*NET task statement)
How this row was scored
Exposure score: 25 out of 100 (18–32 allowing for uncertainty): low exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Drawing software helps, but the measurements and layout come from someone at the scene.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Preparing and serving search and arrest warrants
staying human
This work happens in the physical world: search, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Prepare and serve search and arrest warrants.” (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 same decision, made over and over; work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: The paperwork drafts easily, but only a sworn officer may serve a warrant in person.
The five ratings: output a model can produce 3/4 · needs a body in a room 2/4 · needs an accountable person 3/4 · needs to be trusted in the moment 0/4 · how much data exists 3/4.
Collaborating with representatives from other government and intelligence organizations to share information or coordinating intelligence activities
staying human
The rules require a named, qualified person to answer for representatives, and that person cannot be a piece of software.
importance 4 · Core
Source: “Collaborate with representatives from other government and intelligence organizations to share information or coordinate intelligence activities.” (O*NET task statement)
How this row was scored
Exposure score: 17 out of 100 (10–24 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: someone qualified has to answer for it; the value is that a specific person does it.
The rating behind it: Working with other agencies runs on cleared people and established relationships more than documents.
The five ratings: output a model can produce 2/4 · needs a body in a room 1/4 · needs an accountable person 2/4 · needs to be trusted in the moment 2/4 · how much data exists 1/4.
Notifying command of situation and requesting assistance
staying human
This work happens in the physical world: command of situation, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Notify command of situation and request assistance.” (O*NET task statement)
How this row was scored
Exposure score: 16 out of 100 (9–23 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: the same decision, made over and over; mistakes that are cheap to catch; work that happens in the physical world.
The rating behind it: Radioing a situation report is easy to draft, but it comes from an officer in the field.
The five ratings: output a model can produce 3/4 · needs a body in a room 3/4 · needs an accountable person 0/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Summoning medical help for injured individuals and alert medical personnel to take statements from them
staying human
This work happens in the physical world: medical help, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Summon medical help for injured individuals and alert medical personnel to take statements from them.” (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: The calls themselves are simple, but the officer is at the scene with the injured person.
The five ratings: output a model can produce 3/4 · needs a body in a room 3/4 · needs an accountable person 1/4 · needs to be trusted in the moment 1/4 · how much data exists 3/4.
Obtaining and verifying evidence by interviewing and observing suspects and witnesses or by analyzing records
staying human
This work happens in the physical world: evidence, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Obtain and verify evidence by interviewing and observing suspects and witnesses or by analyzing records.” (O*NET task statement)
How this row was scored
Exposure score: 13 out of 100 (6–20 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: 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: Record analysis is well suited to software, but interviewing and observing people is not.
The five ratings: output a model can produce 2/4 · needs a body in a room 2/4 · needs an accountable person 2/4 · needs to be trusted in the moment 2/4 · how much data exists 2/4.
Processing film and prints from crime or accident scenes
staying human
This work happens in the physical world: film, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Process film and prints from crime or accident scenes.” (O*NET task statement)
How this row was scored
Exposure score: 11 out of 100 (4–18 allowing for uncertainty): minimal 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: Digital handling helps, but processing film and prints from a scene is physical work.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Organizing scene search
staying human
This work happens in the physical world: scene search, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Organize scene search, assigning specific tasks and areas of search to individual officers and obtaining adequate lighting as necessary.” (O*NET task statement)
How this row was scored
Exposure score: 8 out of 100 (1–15 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Assigning search areas and sorting lighting happens at the scene with officers present.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 2/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Gathering intelligence information by field observation
staying human
This work happens in the physical world: intelligence information, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Gather intelligence information by field observation, confidential information sources, or public records.” (O*NET task statement)
How this row was scored
Exposure score: 7 out of 100 (0–14 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: Public records are easy to search, but field observation and informants need a person out there.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 1/4 · needs to be trusted in the moment 2/4 · how much data exists 1/4.
Performing undercover assignments and maintaining surveillance
staying human
This work happens in the physical world: undercover assignments, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Perform undercover assignments and maintain surveillance, including monitoring authorized wiretaps.” (O*NET task statement)
How this row was scored
Exposure score: 5 out of 100 (0–12 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: 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: Wiretap material can be processed by software, but undercover work is a person in the field.
The five ratings: output a model can produce 2/4 · needs a body in a room 3/4 · needs an accountable person 3/4 · needs to be trusted in the moment 2/4 · how much data exists 1/4.
Interviewing survivors, witnesses, suspects and other law enforcement personnel
staying human
This work happens in the physical world: survivors, witnesses, suspects and other law enforcement personnel, in a real place. Software cannot follow it there.
importance 4 · Supplemental
Source: “Interview survivors, witnesses, suspects, and other law enforcement personnel.” (O*NET task statement)
How this row was scored
Exposure score: 3 out of 100 (0–10 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: 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: Getting a witness or suspect to open up depends on a person building trust in the room.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 2/4 · needs to be trusted in the moment 3/4 · how much data exists 1/4.
Questioning individuals or observing persons and establishments to confirm information given to patrol officers
staying human
This work happens in the physical world: individuals, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Question individuals or observe persons and establishments to confirm information given to patrol officers.” (O*NET task statement)
How this row was scored
Exposure score: 3 out of 100 (0–10 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: 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: Confirming information by questioning people and watching premises depends on being there.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 2/4 · needs to be trusted in the moment 2/4 · how much data exists 1/4.
Interviewing interrogate or interact with witnesses or crime suspects to collect human intelligence
staying human
This work happens in the physical world: interrogate, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Interview, interrogate, or interact with witnesses or crime suspects to collect human intelligence.” (O*NET task statement)
How this row was scored
Exposure score: 2 out of 100 (0–9 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: 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: Interrogation depends on a sworn officer reading a person face to face under legal rules.
The five ratings: output a model can produce 1/4 · needs a body in a room 3/4 · needs an accountable person 3/4 · needs to be trusted in the moment 3/4 · how much data exists 1/4.
Photographing crime or accident scenes for evidence records
staying human
This work happens in the physical world: crime, in a real place. Software cannot follow it there.
importance 5 · Core
Source: “Photograph crime or accident scenes for evidence records.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Photographing a crime scene requires a camera at that scene.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Dusting selected areas of crime scene and lifting latent fingerprints
staying human
This work happens in the physical world: selected areas of crime scene, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Dust selected areas of crime scene and lift latent fingerprints, adhering to proper preservation procedures.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Dusting and lifting fingerprints is careful physical work at the scene.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Looking for trace evidence, such as fingerprints, hairs, fibers or shoe impressions
staying human
This work happens in the physical world: trace evidence, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Look for trace evidence, such as fingerprints, hairs, fibers, or shoe impressions, using alternative light sources when necessary.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Looking for hairs, fibres and prints with light sources is physical searching.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Packaging, storing and retrieving evidence
staying human
This work happens in the physical world: evidence, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Package, store and retrieve evidence.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Packaging, storing and retrieving physical evidence means handling the items.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Performing emergency work during off-hours
staying human
This work happens in the physical world: emergency work during off-hours, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Perform emergency work during off-hours.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–7 allowing for uncertainty): minimal exposure, medium confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Emergency call-out work means going to the incident in person.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 1/4 · how much data exists 1/4.
Operating cameras, radios or other surveillance equipment to intercept communications or documenting activities
staying human
This work happens in the physical world: cameras, radios or other surveillance equipment, in a real place. Software cannot follow it there.
importance 3 · Supplemental
Source: “Operate cameras, radios, or other surveillance equipment to intercept communications or document activities.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Operating cameras and surveillance radios is physical equipment work in the field.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Analyzing and processing evidence at crime scenes
staying human
This work happens in the physical world: evidence, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Analyze and process evidence at crime scenes, during autopsies, or in the laboratory, wearing protective equipment and using powders and chemicals.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Processing evidence with powders and chemicals is hands-on laboratory and scene work.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Checking victims for signs of life
staying human
This work happens in the physical world: victims, in a real place. Software cannot follow it there.
importance 5 · Core
Source: “Check victims for signs of life, such as breathing and pulse.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Checking a person for breathing and a pulse requires hands on them.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.
Securing deceased body and obtaining evidence from it
staying human
This work happens in the physical world: deceased body, in a real place. Software cannot follow it there.
importance 5 · Core
Source: “Secure deceased body and obtain evidence from it, preventing bystanders from tampering with it prior to medical examiner's arrival.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Securing a body and taking evidence from it is physical scene work.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Preserving, processing
staying human
This work happens in the physical world: items of evidence, in a real place. Software cannot follow it there.
importance 5 · Core
Source: “Preserve, process, and analyze items of evidence obtained from crime scenes and suspects, placing them in proper containers and destroying evidence no longer needed.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Bagging, storing and destroying physical evidence means handling the items themselves.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Noting, marking and photographing location of objects found, such as footprints, tire tracks
staying human
This work happens in the physical world: location of objects found, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Note, mark, and photograph location of objects found, such as footprints, tire tracks, bullets and bloodstains, and take measurements of the scene.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Marking, photographing and measuring objects at a scene is physical work there.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 2/4.
Securing persons at scene, keeping witnesses from conversing or leaving the scene before investigators arrive
staying human
This work happens in the physical world: persons, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Secure persons at scene, keeping witnesses from conversing or leaving the scene before investigators arrive.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; the value is that a specific person does it.
The rating behind it: Keeping witnesses apart and on scene requires an officer standing there.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 2/4 · how much data exists 1/4.
Participating or assisting in raids and arrests
staying human
This work happens in the physical world: raids, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Participate or assist in raids and arrests.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Taking part in a raid or arrest is physical action by sworn officers.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 3/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Testifying before grand juries concerning criminal activity investigations
staying human
This work happens in the physical world: before grand juries concerning criminal activity investigations, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Testify before grand juries concerning criminal activity investigations.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it; the value is that a specific person does it.
The rating behind it: Testifying before a grand jury is a sworn person giving evidence in the room.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 4/4 · needs to be trusted in the moment 2/4 · how much data exists 2/4.
Blocking or roping off scene and checking perimeter to ensure that entire scene
staying human
This work happens in the physical world: scene, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Block or rope off scene and check perimeter to ensure that entire scene is secured.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world.
The rating behind it: Roping off a scene and walking the perimeter is physical work.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 1/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Searching for and collecting evidence
staying human
This work happens in the physical world: and collecting evidence, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Search for and collect evidence, such as fingerprints, using investigative equipment.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–7 allowing for uncertainty): minimal exposure, high confidence, and it moved between repeat runs, so the range is widened.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Finding and lifting fingerprints and other traces is close hands-on work at the scene.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 3/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Maintaining surveillance of establishments to obtain identifying information on suspects
staying human
This work happens in the physical world: surveillance of establishments, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Maintain surveillance of establishments to obtain identifying information on suspects.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Watching a premises to identify suspects means being there to watch it.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 0/4 · how much data exists 1/4.
Taking fingerprints
staying human
This work happens in the physical world: fingerprints, in a real place. Software cannot follow it there.
importance 4 · Core
Source: “Take fingerprints.” (O*NET task statement)
How this row was scored
Exposure score: 0 out of 100 (0–4 allowing for uncertainty): minimal exposure, high confidence.
Why it sits in this group: work that happens in the physical world; someone qualified has to answer for it.
The rating behind it: Taking fingerprints means guiding someone’s hands onto a scanner or pad.
The five ratings: output a model can produce 0/4 · needs a body in a room 4/4 · needs an accountable person 2/4 · needs to be trusted in the moment 1/4 · how much data exists 2/4.

What this job pays, and how many people do it

Median pay: $93,790a 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: 114,430in 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: progress of investigation in, a record out. The rows above are exactly that shape: gathering, analyzing, correlate or evaluating information from a variety of resources and recording progress of investigation. What it cannot do is be answerable: court needs a named person the rules will accept, and software cannot be that person. 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. Gathering, analyzing, correlate or evaluating information from a variety of resources is going; testifying in court and presenting evidence is not.

So, given all that: 29% of this job's task weight sits in rows the software is already learning, 9% in rows that change shape rather than disappear, and 62% in rows it is nowhere near. That is the position, measured across 67 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 (testifying in court and presenting evidence) 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 testifying in court and presenting evidence, 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 detectives and criminal investigators (nearest by the work that AI is not taking, not by job title), and none of them survived. The closest was forensic science technicians: only about 8% of its durable work is work you already do, it pays 23.2% less and there are far fewer of those jobs than of yours. And on the numbers you do not need one. This job scores 32/100 here, with only 29% of the task list in the top band, and “prepare charges or responses to charges, or information for court cases, according…” is not work that hands over cleanly. None of them beats deepening what you already have.

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.

Forensic Science Technicians
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already photograph crime or accident scenes for evidence records, and their equivalent is to measure and sketch crime scenes to document evidence. Across both published task lists that is about 8% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 8% of the durable side of that job. That is a different job, not a next step. It is a pay cut, in those words: $72,060 against your $93,790, 23.2% less (OEWS May 2025 (both)). Retraining to earn less is a decision, not advice. And it is a narrow door: about 19,120 of those jobs against 114,430 of yours (OEWS May 2025), 17% as many seats.
Look at that job’s page anyway →
Financial Specialists, All Other
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already testify in court and present evidence, and their equivalent is to testify in court regarding investigation findings. Across both published task lists that is about 6% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 6% of the durable side of that job. That is a different job, not a next step. It is a pay cut, in those words: $81,100 against your $93,790, 13.5% less (OEWS May 2025 (both)). Retraining to earn less is a decision, not advice.
Look at that job’s page anyway →
Police and Sheriff's Patrol Officers
Why it looked obvious: It came up as a near neighbour on the work AI is not taking: you already testify in court and present evidence, and their equivalent is to testify in court to present evidence or act as witness in traffic and…. Across both published task lists that is about 5% of the durable work in that job.
Why I am not recommending it: Almost none of it is work you already do: about 5% of the durable side of that job. That is a different job, not a next step. It is a pay cut, in those words: $76,210 against your $93,790, 18.7% less (OEWS May 2025 (both)). Retraining to earn less is a decision, not advice.
Look at that job’s page anyway →

What I’d stop worrying about

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

The headline number you read somewhere
The big “X% of jobs” figures are about the whole economy, not about you. The number that describes your job is on this page: 29% of its task weight, across 67 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 Police officers (sergeant and below) 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:

Senior police officers

In UK official statistics this job is counted as Police officers (sergeant and below) and Senior police officers. 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 detectives / criminal investigators, 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 29% 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 detectives / criminal investigators. 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 detectives / criminal investigators 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 detectives / criminal investigators 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 detectives / criminal investigators 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 detectives / criminal investigators 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 Detectives and Criminal Investigators?: Not as a job, but it is already doing parts of the work. Across the 67 official task statements scored for Detectives and Criminal Investigators (United States, SOC 33-3021), 29% of the importance-weighted core work is made of tasks today's AI could already do most of. The overall exposure score is 32 out of 100 (range 28–38, band: low). 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 “Detectives and Criminal Investigators” can AI already do?: The highest-scoring tasks in release 2026-q4.1 are: “Evaluate records of communications, such as telephone calls, to plot activity and determine the size and location of criminal groups and members” (81/100, very high); “Gather, analyze, correlate, or evaluate information from a variety of resources, such as law enforcement databases” (81/100, very high); “Link or chart suspects to criminal organizations or events to determine activities and interrelationships” (81/100, 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 “Detectives and Criminal Investigators” stay human?: About 62% 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: “Take fingerprints” (0/100, minimal); “Maintain surveillance of establishments to obtain identifying information on suspects” (0/100, minimal); “Search for and collect evidence, such as fingerprints, using investigative equipment” (0/100, 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 “Detectives and Criminal Investigators” do about AI?: Start from the ledger rather than the headline: 29% of this job's weighted core work is exposed, and roughly 62% 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 Detectives and Criminal Investigators 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 67 scored tasks. The prompt, the rubric, the formula and the full dataset are published at https://futureproof.collab365.com/method and https://futureproof.collab365.com/data/2026-q4.1 under CC BY 4.0.

Where these numbers come from

About the data on this page

One task scored differently between repeat runs, so its range on this page is wider. We would rather show the wobble than hide it.
The match between this job and its counterpart in the other country is partial, so the two pages count slightly different groups of people.
3 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 2026-08-04.
Pay and employment: bls-oews (May 2025 estimates (national_M2025_dl.xlsx))bls-oews (May 2025 estimates (national_M2025_dl.xlsx))

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

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

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

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

Using these figures?

Cite this

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

Plain text

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

BibTeX

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

Permalink

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

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