Government jobs safe from AI: the public-service roles with lower AI exposure
The government jobs with the lowest AI exposure are in the field or on the bench: EMTs and paramedics, firefighters and their supervisors, foresters, police supervisors and judges all sit in the bottom half of Microsoft's 785-occupation ranking and under 60/100 on our risk score, at BLS medians of about $44K to $154K. The paperwork side of government, processing claims, checking tax returns, keeping court records and screening benefit applications, is some of the most exposed work anywhere.
That 12/100 is the average for the roles on this list. What's yours?
Experience moves this number a long way in both directions, and it is the only thing about you we cannot read off this page. Answer two questions and GPT-4o writes you the 3 moves that keep you on the safe side of this, the skills gap for each, and a 90-day plan — free, in about 30 seconds.
How long have you been working?
Your own risk score and 3 AI-resistant paths start writing on the next screen. Free, no card, about 30 seconds. Skip the questions.
New · every week
New remote roles in your field — with each one's AI-exposure score — every week
10–15 remote postings from the past week, picked from company career pages and the Himalayas and Jobicy boards. Each shows how exposed the occupation is to AI, from Microsoft's 2025 study of 785 jobs. You apply on the original posting.
Our risk score
12–58/100
Microsoft exposure
p2–p49
BLS median pay
$44,470–$153,990
Tax examiners and eligibility interviewers rank in the top fifth of Microsoft's occupations, and agencies under budget pressure are already piloting AI for exactly those queues. The field and bench roles are different because they combine physical presence with legal authority. That's a pairing generative AI can assist but not take over, because someone accountable has to make the call and be there to make it.
This list is short on purpose. We started from a named set of 44 public-service occupations, the jobs you mostly find in agencies, courts, schools, emergency services and the postal service. Then we kept only the ones that clear both bars. Many classic civil-service jobs didn't, and you can see them further down.
How this list was built
A named list of 44 public-service occupations (emergency services, law enforcement, courts, postal service, forestry and conservation, inspection, public schools and social services). Kept only if Microsoft's AI applicability percentile is 50 or lower and our risk score is under 60, then ranked by our risk score, lowest first. Some of these jobs also exist in the private sector. The list is about the work, not the employer.
Each role shows two different numbers. Our risk score (0–100) is the one on its job page. AI applicability comes from Microsoft's 2025 study of Copilot use across 785 occupations. Lower means less exposure on both. Neither one is a promise about any single job.
The list
| Role | Our risk | Microsoft percentile | BLS median pay |
|---|---|---|---|
| Emergency Medical Technicians | 12/100 | p24 | $44,470 |
| Paramedics | 12/100 | p24 | $60,600 |
| First-Line Supervisors of Firefighting and Prevention Workers | 25/100 | p2 | $93,530 |
| First-Line Supervisors of Police and Detectives | 25/100 | p26 | $106,040 |
| Postmasters and Mail Superintendents | 30/100 | p36 | $96,660 |
| Foresters | 32/100 | p20 | $76,400 |
| Forest Fire Inspectors and Prevention Specialists | 35/100 | p11 | $56,870 |
| Firefighters | 36/100 | p20 | $59,280 |
| Judges, Magistrate Judges, and Magistrates | 37/100 | p23 | $153,990 |
| Administrative Law Judges, Adjudicators, and Hearing Officers | 37/100 | p49 | $117,860 |
| Fire Inspectors and Investigators | 42/100 | p38 | $75,920 |
| Fish and Game Wardens | 58/100 | p43 | $74,060 |
1.Emergency Medical Technicians
Risk: 12/100- AI applicability 0.08: the 24th percentile of 785 occupations
- BLS median pay: $44,470
Assess injuries and illnesses and administer basic emergency medical care.
2.Paramedics
Risk: 12/100- AI applicability 0.08: the 24th percentile of 785 occupations
- BLS median pay: $60,600
Administer basic or advanced emergency medical care and assess injuries and illnesses.
- AI applicability 0.01: the 2nd percentile of 785 occupations
- BLS median pay: $93,530
Directly supervise and coordinate activities of workers engaged in firefighting and fire prevention and control.
A task our analysis keeps with a person: Assign firefighters to jobs at strategic locations to facilitate rescue of persons and maximize application of extinguishing agents.
Full breakdown for First-Line Supervisors of Firefighting and Prevention Workers →
4.First-Line Supervisors of Police and Detectives
Risk: 25/100- AI applicability 0.08: the 26th percentile of 785 occupations
- BLS median pay: $106,040
Directly supervise and coordinate activities of members of police force.
A task our analysis keeps with a person: Supervise and coordinate the investigation of criminal cases, offering guidance and expertise to investigators, and ensuring that procedures are conducted in accordance with laws and regulations.
Full breakdown for First-Line Supervisors of Police and Detectives →
5.Postmasters and Mail Superintendents
Risk: 30/100- AI applicability 0.11: the 36th percentile of 785 occupations
- BLS median pay: $96,660
Plan, direct, or coordinate operational, administrative, management, and support services of a U.S.
A task our analysis keeps with a person: Hire and train employees, and evaluate their performance.
6.Foresters
Risk: 32/100- AI applicability 0.07: the 20th percentile of 785 occupations
- BLS median pay: $76,400
Manage public and private forested lands for economic, recreational, and conservation purposes.
A task our analysis keeps with a person: Negotiate terms and conditions of agreements and contracts for forest harvesting, forest management and leasing of forest lands.
7.Forest Fire Inspectors and Prevention Specialists
Risk: 35/100- AI applicability 0.05: the 11th percentile of 785 occupations
- BLS median pay: $56,870
Enforce fire regulations, inspect forest for fire hazards, and recommend forest fire prevention or control measures.
A task our analysis keeps with a person: Conduct wildland firefighting training.
Full breakdown for Forest Fire Inspectors and Prevention Specialists →
8.Firefighters
Risk: 36/100- AI applicability 0.07: the 20th percentile of 785 occupations
- BLS median pay: $59,280
Control and extinguish fires or respond to emergency situations where life, property, or the environment is at risk.
A task our analysis keeps with a person: Rescue survivors from burning buildings, accident sites, and water hazards.
Full breakdown for Firefighters → · Will AI replace them?
9.Judges, Magistrate Judges, and Magistrates
Risk: 37/100- AI applicability 0.08: the 23rd percentile of 785 occupations
- BLS median pay: $153,990
Arbitrate, advise, adjudicate, or administer justice in a court of law.
A task our analysis keeps with a person: Instruct juries on applicable laws, direct juries to deduce the facts from the evidence presented, and hear their verdicts.
Full breakdown for Judges, Magistrate Judges, and Magistrates →
- AI applicability 0.14: the 49th percentile of 785 occupations
- BLS median pay: $117,860
Conduct hearings to recommend or make decisions on claims concerning government programs or other government-related matters.
A task our analysis keeps with a person: Complex decision-making requiring ethical judgment
Full breakdown for Administrative Law Judges, Adjudicators, and Hearing Officers →
11.Fire Inspectors and Investigators
Risk: 42/100- AI applicability 0.11: the 38th percentile of 785 occupations
- BLS median pay: $75,920
Inspect buildings to detect fire hazards and enforce local ordinances and state laws, or investigate and gather facts to determine cause of fires and explosions.
A task our analysis keeps with a person: Package collected pieces of evidence in securely closed containers, such as bags, crates, or boxes, to protect them.
12.Fish and Game Wardens
Risk: 58/100- AI applicability 0.13: the 43rd percentile of 785 occupations
- BLS median pay: $74,060
Patrol assigned area to prevent fish and game law violations.
A task our analysis keeps with a person: Protect and preserve native wildlife, plants, or ecosystems.
Authority plus presence
EMTs and paramedics come out lowest on our score. Fire and police supervisors come next, then forestry and fire prevention. The common thread is that the job holder is legally on the hook for a decision made on the spot, often in physical danger. Judges and administrative law judges make the list for the same reason, with the bench in place of the field.
If you're weighing a public-sector move, the stable part of government isn't the desk. It's the badge, the bench and the field post. Desk roles still exist, but expect them to be redesigned around AI-assisted case handling, with fewer people reviewing more cases each.
Classic civil-service jobs with high AI exposure
These public-sector roles fail one or both measures:
- Tax Examiners and Collectors, and Revenue Agents: risk 93/100 · AI applicability 0.34: the 96th percentile of 785 occupations · BLS median $62,370
- Eligibility Interviewers, Government Programs: risk 89/100 · AI applicability 0.28: the 85th percentile of 785 occupations · BLS median not published
- Urban and Regional Planners: risk 36/100 · AI applicability 0.33: the 94th percentile of 785 occupations · BLS median $89,320
Not on the list? Check your own job
We score 1,000+ occupations. Search yours to see both numbers.
New · every week
New remote roles in your field — with each one's AI-exposure score — every week
10–15 remote postings from the past week, picked from company career pages and the Himalayas and Jobicy boards. Each shows how exposed the occupation is to AI, from Microsoft's 2025 study of 785 jobs. You apply on the original posting.
Frequently asked questions
Which government jobs are least exposed to AI?
On our data, EMTs, paramedics, fire and police supervisors, foresters and forest-fire inspectors score lowest among public-service roles that also have below-median generative-AI exposure on Microsoft's 2025 study.
Will AI replace government clerical jobs?
They're the most exposed part of government. Tax examiners score 93/100 on our risk score and sit in the top 5% of Microsoft's ranking. Eligibility interviewers for government programs are also high on both measures.
Is a government job safer than a private one?
Not automatically. Civil-service rules can slow layoffs, but the work itself is what's exposed. A benefits processor in an agency and a claims processor at an insurer face similar AI exposure.
Related
Sources and method
- Risk score: ReplacedByAI's 0–100 score from O*NET task, skill and work-activity data, weighted 40% on the Frey & Osborne computerisation probabilities. Full methodology.
- AI applicability: Kiran Tomlinson, Sonia Jaffe, Will Wang, Scott Counts and Siddharth Suri, “Working with AI: Measuring the Applicability of Generative AI to Occupations” (Microsoft, 2025). Data from microsoft/working-with-ai, licensed CC BY 4.0. Percentiles are our calculation, among its 785occupations. Joined on the 6-digit SOC code, so O*NET specialties share their group's score.
- Pay: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, national annual median wage, May 2025.
- Filters: Microsoft percentile 50 or lower and our risk score under 60. “All other” catch-all occupations are excluded.