Natural Sciences Managers
In short: today's best AI models could do about 31% of this job's working time (projected from METR's April 2026 measurement and Epoch's November 2026 index) if the work were set up for them, and about 65% by the end of 2028 if progress keeps its long-run pace. The conversations, in-person and hands-on work, about 34% of the time, stays with people.
Plan, direct, or coordinate activities in such fields as life sciences, physical sciences, mathematics, statistics, and research and development in these fields.
108,690 people do this job in the US. $167,220 a year.
Reviewed on September 30, 2026: what each task is made of, and what handing it over would ask.
A typical 40-hour week
- AI could take on0 h
- A person checks AI's work12 h
- Desk work AI can't do yet14 h
- Stays with people14 h
Natural Sciences Managers, now: of a 40-hour week, AI could take on 0 hours, 12 hours become checking its work, 14 hours are desk work it can't do yet, and 14 hours stay with people.
The largest pieces: Review activities and reports (about 2 h, a person checks); Design phases of analysis and testing (about 2 h, a person checks); Direct research and production (about 1 h, a person checks).
Estimates for a typical week of this job, if progress keeps its long-run pace, counting work the best models finish four times in five. Every task is below.
of the working time. Now is projected from METR's April 2026 measurement and Epoch's November 2026 index.
- 65%
- 1%
- 23%
- 4%
- 7%
of the work: Anthropic's measure of what people already bring to Claude at work (2025). It counts help with part of a task, so it can be higher than the share within reach.
On the chosen pace and reliability. Estimate.
Share of the working time within reach, now to 2030
Change assumptionsHide assumptionsLong-run pace, four in five
What each choice means: , , , , . Every pace starts from .
At this date and pace, models finish tasks up to 2.4 work days long four times in five.
What handing it over would take, now
The 31% within reach splits by what a model would need. The dates, pace and reliability follow the controls above. Estimates.
of the working time. The material can be handed to a model and the result handed back.
of the working time. The model has to be connected to the system the work lives in.
is conversation short enough by length. This site leaves conversation with people, so it is not in the total.
How far this has come
With GPT-4, measured in March 2023, 0% of this job's working time was within reach at . With Claude 3.7 Sonnet, measured in February 2025, 0%. With Claude Mythos Preview (early), measured in April 2026, 3%. Today's estimate: 31%.
A backcast: today's task data with the best model METR had measured by each date. It shows how fast the models have moved, not what anyone forecast at the time.
Measured use over time
Of the Claude.ai conversations Anthropic sampled and tied to a work task, 0.053% went to this job's tasks in May 2026, the 231st largest of the 447 jobs that showed up at all; 3 of its 16 tasks showed up. Within each of Anthropic's methods, never across a change: 0.058% to 0.032% (released February and March 2025); 0.033% to 0.028% (August 2025 to February 2026); 0.057% to 0.053% (April to May 2026).
Each release
| Conversations | Share | Rank | Tasks seen |
|---|---|---|---|
| Released February 2025 | 0.058% | 248 of 708 | 6 of 16 |
| Released March 2025 | 0.032% | 346 of 703 | 6 of 16 |
| The week of Aug 4, 2025 (released September 2025) | 0.033% | 292 of 655 | 4 of 16 |
| The week of Nov 13, 2025 (released January 2026) | 0.026% | 354 of 672 | 5 of 16 |
| The week of Feb 5, 2026 (released March 2026) | 0.028% | 350 of 677 | 6 of 16 |
| April 2026 (released June 2026) | 0.057% | 220 of 434 | 3 of 16 |
| May 2026 (released June 2026) | 0.053% | 231 of 447 | 3 of 16 |
Measured, not estimated: Anthropic's samples of Claude.ai conversations, each tied by a model to an O*NET task, from the Anthropic Economic Index. It says where Claude's use goes, not how much of this job's working time it covers, so it is not on the same scale as the reach above. The releases differ in their sample, their classifier and the smallest share they report (the June 2026 file rounds each task to a hundredth of a percent, so a smaller task reads as none), which is why a change of method starts a new line. At least 96% of each older release's task-tied use matched this site's tasks, and all of the newest. Claude.ai only, so use through other tools and other companies' models is not in it. Every job ranked.
Robots and the physical work
Anthropic rates almost none of this job's working time as physical work, so robots do not reach it.
Anthropic's rating, not this site's estimate: Claude Opus 5, searching the web, rated each physical O*NET task by the kind of setting in which a robot it could cite (deployed, sold or demonstrated) does most of the task, weighted by its own estimate of each task's share of the time (What work can robots do?, September 30, 2026, data CC BY 4.0; every job ranked). A robot shown somewhere is not a robot in your building: cost, setup and safety decide that. The share within reach on this page still counts no body work. The robot behind each rated task is under the task below.
What changes for the people in this job
From the task-by-task review, by the end of 2028 on the long-run pace at four in five. It says which parts of the work stay with people and which change, not whether anyone keeps a job: that depends on decisions no data set can see.
What stays with people
34% of the working time: the conversation, in-person and hands-on parts of the tasks.
- Direct research and production, the conversation, 10%
- Conduct own research, the hands-on work, 5%
- Design phases of analysis and testing, the conversation, 4%
- Hire and evaluate staff, the conversation, 3%
- Confer on projects, the conversation, 2%
- Steward resources and habitats, the hands-on work, 2%
What becomes checking the machine
64% of the working time: desk work a model could do by then, in tasks where a mistake costs something or a licensed person signs. A person checks it.
- Conduct own research, costly if wrong, 10%
- Design phases of analysis and testing, costly if wrong, 10%
- Review activities and reports, costly if wrong, 9%
- Set technical goals and plans, costly if wrong, 7%
- Direct research and production, costly if wrong, 6%
- Develop policies and standards, costly if wrong, 6%
What to learn
- Reviewing work you did not do yourself. The tasks under What becomes checking the machine, where a mistake costs something, stay a person's to check: spotting what a draft gets wrong is a skill of its own.
- Briefing a model and checking its drafts. About 65% of the time is documents and data, the part models reach first: writing a clear brief, giving the right material, and checking what comes back.
- The conversations. About 27% of the time is conversation and in-person work, which this site leaves with people: listening, explaining, persuading, calming.
Your own version of this job: read it against your job description. The headline is free; the full personal report, every task with its timeline and what to learn, is $19 (a sample).
The tasks
Sorted by . 16 tasks from .
| Task | ||||
|---|---|---|---|---|
| Plan or direct research, development, or production activities.Mostly conversation with the people doing the work, plus reading and deciding.Done daily by most who do it. | 16% | 1 work day2.5 h to 3.1 work days | Conversation 60%, documents 40% | 21%Desk part half by now |
| Conduct own research in field of expertise.Analysis and writing, with some bench or field work.Done more than yearly by most who do it. AI observed on this task. | 16% | 2.1 work days3.3 h to 1.7 work weeks | Documents 70%, hands-on 30% | 10%Desk part half by Oct 2027 |
| Design or coordinate successive phases of problem analysis, solution proposals, or testing.Phases planned on paper, with conversations.Done more than monthly by most who do it. | 14% | 6.7 h1.7 h to 2.1 work days | Documents 70%, conversation 30% | 43%Desk part half by now |
| Review project activities and prepare and review research, testing, or operational reports.Reports prepared and reviewed.Done daily by most who do it. AI observed on this task. | 9% | 5 h1.3 h to 1.7 work days | Documents 100% | 71%Half by now |
| Determine scientific or technical goals within broad outlines provided by top management and make detailed plans to accomplish these goals.Goals and plans worked out on paper.Done more than monthly by most who do it. | 8% | 1.3 work days2.5 h to 3.1 work days | Documents 100% | 44%Half by Dec 2026 |
| Hire, supervise, or evaluate engineers, technicians, researchers, or other staff.Interviews and reviews in conversation, with the write-ups behind them.Done more than yearly by most who do it. | 6% | 2.5 h50 min to 1 work day | Conversation 60%, documents 40% | 37%Desk part half by now |
| Develop or implement policies, standards, or procedures for the architectural, scientific, or technical work performed to ensure regulatory compliance or operations enhancement.Policies and procedures written.Done more than yearly by most who do it. | 6% | 1.7 work days3.3 h to 1.3 work weeks | Documents 100% | 24%Half by May 2027 |
| Develop innovative technology or train staff for its implementation.Development on paper and training in person.Done more than yearly by most who do it. | 5% | 2.1 work days3.3 h to 1.3 work weeks | Documents 60%, in person 40% | 11%Desk part half by Jul 2027 |
| Prepare and administer budgets, approve and review expenditures, and prepare financial reports.Budgets worked out on paper and tracked in the system.Done more than yearly by most who do it. | 5% | 6.7 h1.7 h to 2.5 work days | Documents 80%, systems 20% | 55%Half by now |
| Provide for stewardship of plant or animal resources or habitats, studying land use, monitoring animal populations, or providing shelter, resources, or medical treatment for animals.Land and populations studied on paper and in the field.Done daily by most who do it. | 4% | 1 work day1.7 h to 4.2 work days | Documents 50%, hands-on 50% | 16%Desk part half by Mar 2027 |
| Confer with scientists, engineers, regulators, or others to plan or review projects or to provide technical assistance.Conversations with scientists, engineers and regulators, with the plans.Done more than weekly by most who do it. | 3% | 1.7 h25 min to 5 h | Conversation 70%, documents 30% | 28%Desk part half by now |
| Recruit personnel or oversee the development or maintenance of staff competence.Conversations with candidates and staff, with the plans and records.Done more than yearly by most who do it. | 3% | 4.2 h1.3 h to 1.6 work days | Conversation 60%, documents 40% | 32%Desk part half by now |
Within reach, per task: the share of the task's time whose instances are short enough for a model to finish at the chosen date, pace and reliability. Half by: the date half of that time comes within reach. Made of: how the task's time splits among five kinds of doing. Only the documents and systems parts can count as within reach, so a task that is 30% documents can be at most 30% within reach.
Jobs that share skills with this one
From O*NET's related occupations. Within reach now and by the end of 2028, long-run pace, four in five.
Where this job works
The industry groups that employ the most of it (BLS, May 2025).
Where the most of these jobs are
The metro areas with the most people in it (BLS, May 2025).
Questions about this job
- How much of the work of Natural Sciences Managers is within reach of AI now?
- An estimated 31% of the working time is within reach of AI models now and 65% by the end of 2028, on METR's long-run pace at four in five (57% to 66% across the three paces the site shows). Within reach means a model could finish work that long at that reliability. It is not a forecast of jobs lost or of savings.
- Which tasks of Natural Sciences Managers come within reach first?
- Weighted by the time each takes, the tasks with the most within reach now (projected from METR's April 2026 measurement and Epoch's November 2026 index) are: Review activities and reports (71% of the task's time now, 100% by the end of 2028); Design phases of analysis and testing (43% of the task's time now, 70% by the end of 2028); and Direct research and production (21% of the task's time now, 40% by the end of 2028).
- Which tasks of Natural Sciences Managers stay with people?
- Every task has a desk part, so each can come partly within reach; the conversation, in-person and hands-on parts of each task stay with people, about 34% of the working time.
- What should Natural Sciences Managers learn as AI takes on more of the work?
- By the job's own task-by-task review, the first things to learn are reviewing work you did not do yourself; briefing a model and checking its drafts; and the conversations. By the end of 2028 on the long-run pace at four in five, about 34% of the working time stays with people, 64% becomes checking a model's work and 0% could be handed over with lighter checks. It says which parts of the work change, never whether anyone keeps a job.
- Do people already use AI for the work of Natural Sciences Managers?
- In the Claude.ai conversations Anthropic sampled for May 2026 and tied to a work task, 0.053% went to this job's tasks, the 231st largest share of the 447 jobs that showed up; 3 of its 16 tasks appeared. Anthropic's labor market measure puts the share of the job's tasks seen in real Claude use at work at 6%. These are measures of use, not of how much of the work AI could finish, and they cover Claude only.
- Can robots do the physical work of Natural Sciences Managers?
- Anthropic's robot exposure rating (September 2026) finds almost none of the working time is physical work, so robots do not reach it.
- How is this estimated?
- From the 16 tasks O*NET lists for the occupation, each with an estimated length checked against GDPval and a share of the working time, reviewed one task at a time on September 30, 2026 for what each is made of (documents, business systems, conversation, in person, hands-on), and METR's measurements of how long a task the best AI models can finish, carried forward to the date from the latest, Claude Mythos Preview (early) in April 2026. Data checked November 10, 2026.
This job in your state or metro area
Embed these figures on your site
A small card with the share of working time within reach now and at the end of 2027, 2028 and 2030, the date and a link back. Free under CC BY 4.0; the card carries the credit.
Your own version of this job
The page above is the occupation. Your job is your title and what you actually do: give both, and they are read against these tasks in your browser, nothing sent anywhere. The headline is free: how much of your working time is within reach now and by the end of 2028, and how many of the tasks your description names.
The full personal report is $19, to print or save as a PDF: every task with when it comes within reach, which ones your description names, the parts of your description the task list does not cover, what stays with you, what becomes checking the machine and what to learn.