Carpenters
In short: today's best AI models could do about 15% 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 the same by the end of 2028 if progress keeps its long-run pace. The conversations, in-person and hands-on work, about 85% of the time, stays with people.
Construct, erect, install, or repair structures and fixtures made of wood and comparable materials, such as concrete forms; building frameworks, including partitions, joists, studding, and rafters; and wood stairways, window and door frames, and hardwood floors. May also install cabinets, siding, drywall, and batt or roll insulation. Includes brattice builders who build doors or brattices (ventilation walls or partitions) in underground passageways.
670,090 people do this job in the US. $60,580 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 on3 h
- A person checks AI's work3 h
- Desk work AI can't do yet0 h
- Stays with people34 h
Carpenters, now: of a 40-hour week, AI could take on 3 hours, 3 hours become checking its work, 0 hours are desk work it can't do yet, and 34 hours stay with people.
The largest pieces: Keep job records and schedule the crew (about 2 h); Keep records and write progress reports (about 1 h); Study blueprints to lay out the project (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.
- 10%
- 5%
- 1%
- 1%
- 83%
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 15% 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, 13%. Today's estimate: 15%.
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, 0.032% went to this job's tasks in May 2026, the 260th largest of the 447 jobs that showed up at all; 1 of its 29 tasks showed up. Within each of Anthropic's methods, never across a change: 0.013% to 0.015% (released February and March 2025); 0.013% to 0.0087% (August 2025 to February 2026); 0.023% to 0.032% (April to May 2026).
Each release
| Conversations | Share | Rank | Tasks seen |
|---|---|---|---|
| Released February 2025 | 0.013% | 502 of 708 | 2 of 29 |
| Released March 2025 | 0.015% | 459 of 703 | 2 of 29 |
| The week of Aug 4, 2025 (released September 2025) | 0.013% | 428 of 655 | 4 of 29 |
| The week of Nov 13, 2025 (released January 2026) | 0.006% | 552 of 672 | 3 of 29 |
| The week of Feb 5, 2026 (released March 2026) | 0.0087% | 513 of 677 | 4 of 29 |
| April 2026 (released June 2026) | 0.023% | 293 of 434 | 1 of 29 |
| May 2026 (released June 2026) | 0.032% | 260 of 447 | 1 of 29 |
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 88% of this job's working time as physical work. For 68% of the working time, a robot has been shown doing most of the task somewhere: 67% in a purpose-built cell, 1% in a structured facility. 0.70 of 3, the 261st highest of the 923 jobs.
- Not physical 12%
- No robot shown 20%
- In a purpose-built cell 67%
- In a structured facility 1%
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
85% of the working time: the conversation, in-person and hands-on parts of the tasks.
- Build and repair cabinets and fixtures, the hands-on work, 14%
- Install windows, floors and trim, the hands-on work, 9%
- Repair or replace damaged sections, the hands-on work, 5%
- Work with or remove hazardous material, the hands-on work, 5%
- Lay wood floors, the hands-on work, 5%
- Build hauling sleds from timber, the hands-on work, 5%
What becomes checking the machine
7% 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. Another 8% is desk work that is small if wrong, which can be handed over with lighter checks.
- Study blueprints to lay out the project, costly if wrong, 3%
- Prepare cost estimates, costly if wrong, 2%
- Select and order lumber and materials, costly if wrong, 2%
What to learn
- The hands-on craft. About 83% of the time is hands-on work, which no model here does; robots are rated separately on this page.
- Judgment where errors are severe. Some tasks are severe if wrong. Knowing when not to trust an output matters more there than speed.
- 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 or sign: spotting what a draft gets wrong is a skill of its own.
- Signing off under your license. Some tasks need a licensed, certified or sworn person to do or sign them. The license, and the judgment it stands for, stays with the person.
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 . 29 tasks from .
| Task | ||||
|---|---|---|---|---|
| Build or repair cabinets, doors, frameworks, floors, or other wooden fixtures used in buildings, using woodworking machines, carpenter's hand tools, or power tools.Woodworking by hand and machine.Done more than monthly by most who do it. Robots: in a purpose-built cell, Randek ZeroLabor Robotic System (framing, sub-assembly, sheathing) (sold). Anthropic's rating. | 14% | 2.5 h50 min to 6.7 h | Hands-on 100% | Stays with people |
| Install structures or fixtures, such as windows, frames, floorings, trim, or hardware, using carpenters' hand or power tools.Hands-on work with tools.Done more than weekly by most who do it. Robots: no robot shown. Anthropic's rating. | 9% | 1.3 h25 min to 3.3 h | Hands-on 100% | Stays with people |
| Remove damaged or defective parts or sections of structures and repair or replace, using hand tools.Hands-on work with tools.Done daily by most who do it. Robots: no robot shown. Anthropic's rating. | 5% | 1.3 h25 min to 3.3 h | Hands-on 100% | Stays with people |
| Work with or remove hazardous material.Hands-on work that takes certified training.Done yearly or less by most who do it.Robots: in a purpose-built cell, Hundegger Robot-Drive / Speed-Cut CNC joinery machines (Hans Hundegger AG) (sold). Anthropic's rating. | 5% | 3.3 h1.7 h to 1.7 work days | Hands-on 100% | Stays with people |
| Cover subfloors with building paper to keep out moisture and lay hardwood, parquet, or wood-strip-block floors by nailing floors to subfloor or cementing them to mastic or asphalt base.Hands-on work.Done more than monthly by most who do it. Robots: in a purpose-built cell, Randek AutoFloor floor/roof element production system (sold). Anthropic's rating. | 5% | 5 h1.7 h to 1.7 work days | Hands-on 100% | Stays with people |
| Maintain job records and schedule work crew.Job records and the crew schedule.Done daily by most who do it. | 5% | 38 min17 min to 1.7 h | Systems 60%, documents 40% | 100%Half by now |
| Build sleds from logs and timbers for use in hauling camp buildings and machinery through wooded areas.Heavy carpentry.Done more than yearly by most who do it.Robots: in a purpose-built cell, Krüsi CM-40 log home milling machine (CNC-controlled option) (sold). Anthropic's rating. | 5% | 1.3 work days6.7 h to 2.5 work days | Hands-on 100% | Stays with people |
| Measure and mark cutting lines on materials, using a ruler, pencil, chalk, and marking gauge.Hands-on work.Done daily by most who do it. Robots: in a purpose-built cell, Hundegger SC/K2 joinery machines and SPM-2 Speed Panel Machine (Hans Hundegger AG) (sold). Anthropic's rating. | 4% | 17 min8 min to 38 min | Hands-on 100% | Stays with people |
| Apply shock-absorbing, sound-deadening, or decorative paneling to ceilings or walls.Hands-on work.Done yearly or less by most who do it.Robots: in a purpose-built cell, Randek ZeroLabor robotic sheathing system (sold). Anthropic's rating. | 4% | 5.8 h3.3 h to 1.7 work days | Hands-on 100% | Stays with people |
| Study specifications in blueprints, sketches, or building plans to prepare project layout and determine dimensions and materials required.Reading the drawings and working out dimensions and materials.Done more than monthly by most who do it. | 4% | 1.1 h25 min to 2.5 h | Documents 70%, in person 30% | 70%Desk part half by now |
| Finish surfaces of woodwork or wallboard in houses or buildings, using paint, hand tools, or paneling.Hands-on finishing.Done more than yearly by most who do it. Robots: in a purpose-built cell, Okibo EG7 (deployed). Anthropic's rating. | 4% | 1.7 h38 min to 5 h | Hands-on 100% | Stays with people |
| Follow established safety rules and regulations and maintain a safe and clean environment.Working safely while doing the carpentry.Done daily by most who do it. Robots: in a purpose-built cell, Boston Dynamics Spot (Skanska/Autodesk housekeeping pilot) (demonstrated). Anthropic's rating. | 3% | 25 min8 min to 50 min | Hands-on 100% | Stays with people |
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.
- Drywall and Ceiling Tile Installers5% now, 5% by 2028
- Structural Iron and Steel Workers3% now, 3% by 2028
- Brickmasons and Blockmasons8% now, 8% by 2028
- Sheet Metal Workers18% now, 18% by 2028
- Structural Metal Fabricators and Fitters4% now, 4% by 2028
- Cabinetmakers and Bench Carpenters18% now, 19% by 2028
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 Carpenters is within reach of AI now?
- An estimated 15% of the working time is within reach of AI models now and 15% by the end of 2028, on METR's long-run pace at four in five. 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 Carpenters 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: Keep job records and schedule the crew (100% of the task's time now); Keep records and write progress reports (100% of the task's time now); and Study blueprints to lay out the project (70% of the task's time now).
- Which tasks of Carpenters stay with people?
- Tasks that are people work or body work are not counted as within reach at any date. The largest: Build and repair cabinets and fixtures; Install windows, floors and trim; and Repair or replace damaged sections. Together such tasks are about 83% of the working time, and the people and body parts of the other tasks stay with people too.
- What should Carpenters learn as AI takes on more of the work?
- By the job's own task-by-task review, the first things to learn are the hands-on craft; judgment where errors are severe; and reviewing work you did not do yourself. By the end of 2028 on the long-run pace at four in five, about 85% of the working time stays with people, 7% becomes checking a model's work and 8% 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 Carpenters?
- In the Claude.ai conversations Anthropic sampled for May 2026 and tied to a work task, 0.032% went to this job's tasks, the 260th largest share of the 447 jobs that showed up; 1 of its 29 tasks appeared. Anthropic's labor market measure puts the share of the job's tasks seen in real Claude use at work at 0%. 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 Carpenters?
- Anthropic's robot exposure rating (September 2026) finds 88% of the working time is physical work, and that a robot has been shown doing most of the task, somewhere, for 68% of the working time, 1% in a structured facility. Its index is 0.70 of 3. One example: Hilti Jaibot for "bore boltholes". A robot shown somewhere is not one in every workplace; this site's own estimates still count no physical work as within reach.
- How is this estimated?
- From the 29 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.