SAFE
Mixing drinks, pulling drafts, restocking coolers, cutting fruit, reading a crowded room and defusing a drunk customer are physical, social, real-time tasks that language models cannot touch and current robotics handles only in novelty kiosks with fixed menus. What AI does erode is the paperwork edge: inventory counts, pour-cost analysis, cocktail menu drafting, and scheduling. The bigger threats to this job are self-pour taps, tablet ordering, and tip-economics — not an AI that can hold a bar.
Dipped in 2020, then grew past where it started.
Median pay $23,680 → $34,340 +16.0% in real terms
This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.
So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing. The marked year is 2020.
BLS projection, 2024–2034
+5.9% 756,700 → 801,500 on the projections basis
Hard to automate, and growing
The work resists current AI and the BLS projects +5.9% more of these jobs by 2034. Note that safe does not mean well paid — several of the fastest-growing resistant occupations are among the lowest paid on the register.
Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.
~129,600 openings a year on average, including replacing people who leave.
BarmanBarkeepBarmaidBarkeeperBartenderBar TenderMixologistBar CaptainDrink MixerBar AttendantEvent BartenderBanquet BartenderService BartenderTaproom AttendantCatering BartenderRestaurant Bartender
Holding it up: embodiment . Weakest point: liability shield .
Tasks largely resist digitisation A bartender's shift is building four different drinks at once from muddled mint, chilled glassware and hand-cut citrus while tracking six open tabs and the guy at the end who's had enough — the only pieces software touches are the par-level sheet and the pour-cost spreadsheet, which is why this sits at 16 and not 19.
Hands-on in uncontrolled environments You are standing on rubber mats for eight hours, hauling kegs and ice bins, wiping spills, changing CO2 lines, and reaching across a wet crowded well that no two nights configure the same way — the environment is unpredictable enough that robot arms only work behind a rail with a fixed 12-drink menu.
Certification preferred, not legally required You hold a TIPS or state alcohol-server card and dram-shop statutes can put liability on you for over-serving, but the licence that actually matters is the establishment's liquor licence held by the owner — your certification is a day-long course anyone can pass, so it blocks nothing structurally.
Meaningful discretion Cutting someone off, spotting a fake ID, deciding when a table's argument becomes a door situation, and calling a cab are calls you make in seconds with real legal and safety consequences, but they run against house policy and posted ID rules rather than open-ended professional judgment.
The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.
Your task mix speaks to task resistance (16/20 here) — how much of the day's work current AI already does. That is the dimension the boxes above are about.
It cannot move the other three. Liability shield (8/20) is whether the law requires a licensed human to sign. Trust premium (15/20) is whether buyers specifically pay for a person. Judgment and accountability (11/20) is whether the role exists to own consequential calls. Those are facts about the occupation's standing, not about which tasks are in your week — a paralegal who does only trial exhibits still holds no licence. Together they are 34 of this occupation's 68 points (50%).
Embodiment (18/20) is also a property of the work rather than the worker, but we don't tag individual tasks as physical or not, so the picker can't tell you anything about it. That's a limit of this tool, not a claim.
Did we get the list right? Tell us what's missing — the tasks are written from the outside, and you're reading this from the inside.
The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 81/100, still SAFE.
State ABC boards moving from voluntary to mandatory certified server training with personal liability — e.g. states adopting mandatory RBS/TIPS-style certification (California's RBS program since 2022, Washington MAST) plus dram-shop statutes that name the individual pourer, not just the licensee. Extension of mandatory certification to the remaining ~20 states without it, and any rule that self-pour taps and tablet ordering require a certified human to authorize each additional pour, would make a licensed human the legally required checkpoint on intoxication calls.
Task-mix shift as self-pour walls and tablet ordering absorb the routine beer-and-well-drink tier, leaving crowd-reading, intoxication assessment, conflict de-escalation and bespoke build as the residual job. Note this shrinks headcount while raising per-role resistance — a smaller job that is harder to automate.
Dram-shop and social-host liability litigation increasingly turning on the specific cut-off decision, plus insurer requirements (liquor liability carriers demanding documented refusal-of-service logs signed by the bartender on shift) that formalize the 'who decided to stop serving' call as an auditable, attributable judgment rather than informal discretion.
Continued growth of the craft cocktail and omakase-bar segment where the bartender is the product and menus are bespoke — plus any consumer backlash making 'staffed bar, no QR ordering' an advertised feature. This is a segment effect: it raises the premium for maybe the top tier of venues, not for the high-volume chain and stadium bars where most of the 756k work.
The limit. The realistic ceiling is roughly high-70s. The binding constraint is not AI capability but labor economics: self-pour, tablet ordering and tip-credit changes reduce the number of bartender slots without any model needing to mix a drink. A higher liability shield protects the role's content while headcount still falls.
| New York-Newark-Jersey City, NY-NJ | 38,510 | $61,220 +78% |
| Chicago-Naperville-Elgin, IL-IN | 28,750 | $31,200 -9% |
| Los Angeles-Long Beach-Anaheim, CA | 25,260 | $35,510 +3% |
| Dallas-Fort Worth-Arlington, TX | 18,300 | $24,170 -30% |
| Minneapolis-St. Paul-Bloomington, MN-WI | 15,370 | $27,600 -20% |
| Boston-Cambridge-Newton, MA-NH | 14,840 | $37,760 +10% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 14,560 | $48,030 +40% |
| Miami-Fort Lauderdale-West Palm Beach, FL | 14,510 | $34,700 +1% |
| Kahului-Wailuku, HI | 800 | $86,280 +151% |
| Urban Honolulu, HI | 2,130 | $75,340 +119% |
| New York-Newark-Jersey City, NY-NJ | 38,510 | $61,220 +78% |
We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.
That is worth saying out loud next to a score of 68. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.
Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.
Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.