Security Guards
In short: today's best AI models could do about 14% 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 86% of the time, stays with people.
Guard, patrol, or monitor premises to prevent theft, violence, or infractions of rules. May operate x-ray and metal detector equipment.
1,283,470 people do this job in the US. $38,020 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 on1 h
- A person checks AI's work5 h
- Desk work AI can't do yet0 h
- Stays with people34 h
Security Guards, now: of a 40-hour week, AI could take on 1 hours, 5 hours become checking its work, 0 hours are desk work it can't do yet, and 34 hours stay with people.
The largest pieces: Write daily activity reports (about 4 h, a person checks); Control who enters and leaves (under an hour, a person checks); Inspect security systems and equipment (under an hour, 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.
- 9%
- 5%
- 4%
- 21%
- 61%
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 14% 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, 1%. With Claude Mythos Preview (early), measured in April 2026, 14%. Today's estimate: 14%.
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, no reportable share went to this job's tasks in May 2026; 0 of its 14 tasks showed up. Within each of Anthropic's methods, never across a change: 0.0078% to 0.0065% (released February and March 2025); 0.0023% to none (August 2025 to February 2026).
Each release
| Conversations | Share | Rank | Tasks seen |
|---|---|---|---|
| Released February 2025 | 0.0078% | 559 of 708 | 2 of 14 |
| Released March 2025 | 0.0065% | 553 of 703 | 2 of 14 |
| The week of Aug 4, 2025 (released September 2025) | 0.0023% | 621 of 655 | 1 of 14 |
| The week of Nov 13, 2025 (released January 2026) | none | n/a | 0 of 14 |
| The week of Feb 5, 2026 (released March 2026) | none | n/a | 0 of 14 |
| April 2026 (released June 2026) | none | n/a | 0 of 14 |
| May 2026 (released June 2026) | none | n/a | 0 of 14 |
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 61% of this job's working time as physical work. For 56% of the working time, a robot has been shown doing most of the task somewhere: 10% in a purpose-built cell, 46% in a structured facility. 1.02 of 3, the 107th highest of the 923 jobs.
- Not physical 39%
- No robot shown 5%
- In a purpose-built cell 10%
- In a structured facility 46%
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
86% of the working time: the conversation, in-person and hands-on parts of the tasks.
- Circulate to keep order, the hands-on work, 21%
- Patrol premises, the hands-on work, 15%
- Escort or drive people, the hands-on work, 9%
- Control who enters and leaves, the in-person work, 9%
- Answer alarms and investigate, the hands-on work, 6%
- Inspect security systems and equipment, the hands-on work, 6%
What becomes checking the machine
13% 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 1% is desk work that is small if wrong, which can be handed over with lighter checks.
- Write daily activity reports, costly if wrong, 9%
- Control who enters and leaves, costly if wrong, 2%
- Inspect security systems and equipment, costly if wrong, 1%
What to learn
- The hands-on craft. About 61% 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.
- Dealing with people outside the organization. Customers, patients, suppliers or the public: trust with them is built by a person.
- The conversations. About 25% of the time is conversation and in-person work, which this site leaves with people: listening, explaining, persuading, calming.
- 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.
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 . 14 tasks from . One has a worked example.
| Task | ||||
|---|---|---|---|---|
| Circulate among visitors, patrons, or employees to preserve order and protect property.Being present among people.Done hourly or more by most who do it. Robots: in a structured facility, Cobalt Robot (sold). Anthropic's rating. | 21% | 38 min17 min to 1.7 h | Hands-on 60%, in person 40% | Stays with people |
| Patrol industrial or commercial premises to prevent and detect signs of intrusion and ensure security of doors, windows, and gates.Walking or driving the grounds.Done daily by most who do it. Robots: in a structured facility, Cobalt Security Robot (Cobalt Monitoring Intelligence) (sold). Anthropic's rating. | 15% | 38 min17 min to 1.3 h | Hands-on 100% | Stays with people |
| Monitor and authorize entrance and departure of employees, visitors, and other persons to guard against theft and maintain security of premises.Checking people in at the gate or desk, with the log.Done hourly or more by most who do it. | 11% | 17 min8 min to 38 min | In person 60%, systems 20%, hands-on 20% | 20%Desk part half by now |
| Write reports of daily activities and irregularities, such as equipment or property damage, theft, presence of unauthorized persons, or unusual occurrences.A written report of the shift.Done daily by most who do it. See it done: one graded attempt (passes) | 9% | 25 min13 min to 50 min | Documents 100% | 100%Half by now |
| Escort or drive motor vehicle to transport individuals to specified locations or to provide personal protection.Driving and accompanying people.Done yearly or less by most who do it. Robots: in a purpose-built cell, Cobalt Robotics indoor security robot (deployed). Anthropic's rating. | 9% | 50 min25 min to 2.5 h | Hands-on 80%, in person 20% | Stays with people |
| Inspect and adjust security systems, equipment, or machinery to ensure operational use and to detect evidence of tampering.Checking devices in person, with some settings on screen.Done daily by most who do it. Robots: in a structured facility, Cobalt Robot (Cobalt AI) (sold). Anthropic's rating. | 7% | 25 min13 min to 1.3 h | Hands-on 80%, systems 20% | 20%Desk part half by now |
| Answer alarms and investigate disturbances.Going to the spot to see.Done daily by most who do it. Robots: in a structured facility, Cobalt Robot / Cobalt Monitoring Intelligence (deployed). Anthropic's rating. | 6% | 25 min8 min to 1.3 h | Hands-on 100% | Stays with people |
| Warn persons of rule infractions or violations, and apprehend or evict violators from premises, using force when necessary.Confronting people in person, by force if needed.Done daily by most who do it. Robots: no robot shown. Anthropic's rating. | 5% | 25 min8 min to 1.7 h | In person 60%, hands-on 40% | Stays with people |
| Operate detecting devices to screen individuals and prevent passage of prohibited articles into restricted areas.Running the scanner and checking people in person.Done hourly or more by most who do it. Robots: in a purpose-built cell, TSA/DHS S&T Self-Service Screening System (Innovation Checkpoint, Harry Reid International Airport) (demonstrated). Anthropic's rating. | 5% | 17 min8 min to 38 min | Hands-on 70%, in person 30% | Stays with people |
| Answer telephone calls to take messages, answer questions, and provide information during non-business hours or when switchboard is closed.Live calls, with a message taken.Done daily by most who do it. | 4% | 8 min4 min to 17 min | Conversation 70%, systems 30% | 30%Desk part half by now |
| Respond to medical emergencies by administering basic first aid or by obtaining assistance from paramedics.First aid by hand, and a call for paramedics.Done more than monthly by most who do it. Robots: no robot shown. Anthropic's rating. | 3% | 25 min13 min to 1.3 h | Hands-on 80%, conversation 20% | Stays with people |
| Lock doors and gates of entrances and exits to secure buildings.Walking the building to secure it.Done daily by most who do it. Robots: in a purpose-built cell, Verkada Access Control (door schedules with hardwired and wireless electrified locks) (deployed). Anthropic's rating. | 3% | 8 min4 min to 17 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.
- First-Line Supervisors of Security Workers34% now, 42% by 2028
- Transit and Railroad Police21% now, 32% by 2028
- Transportation Security Screeners11% now, 11% by 2028
- Police and Sheriff's Patrol Officers13% now, 14% by 2028
- Parking Enforcement Workers14% now, 14% by 2028
- Public Safety Telecommunicators49% now, 64% 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 Security Guards is within reach of AI now?
- An estimated 14% of the working time is within reach of AI models now and 14% 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 Security Guards 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: Write daily activity reports (100% of the task's time now); Control who enters and leaves (20% of the task's time now); and Inspect security systems and equipment (20% of the task's time now).
- Which tasks of Security Guards stay with people?
- Tasks that are people work or body work are not counted as within reach at any date. The largest: Circulate to keep order; Patrol premises; and Escort or drive people. Together such tasks are about 68% of the working time, and the people and body parts of the other tasks stay with people too.
- What should Security Guards 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 dealing with people outside the organization. By the end of 2028 on the long-run pace at four in five, about 86% of the working time stays with people, 13% becomes checking a model's work and 1% 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 Security Guards?
- In the Claude.ai conversations Anthropic sampled for May 2026, no reportable share went to this job's tasks. 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 Security Guards?
- Anthropic's robot exposure rating (September 2026) finds 61% of the working time is physical work, and that a robot has been shown doing most of the task, somewhere, for 56% of the working time, 46% in a structured facility. Its index is 1.02 of 3. One example: Cobalt Robot for "circulate to keep order". 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 14 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. The Work Bench has 1 worked example of its tasks: a brief, a rubric written beforehand, one attempt by an AI model and its grade.
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.