← Risk register SOC 51-3023 · reviewed 2026-08-11

Slaughterers and Meat Packers

69,950 US workers · median $40,130/yr · Production

EXPOSED

Almost nothing here is screen or text work — knife work on variable carcasses in cold, wet, high-speed plants is exactly what language models cannot touch and what robots still struggle with, since every animal differs in size and bone geometry. The real exposure is mechanical, not conversational: automated primal saws, X-ray-guided cutting, and robotic pork deboning lines are already deployed in Denmark, the Netherlands, and large US pork plants, and each generation pushes further into hand-boning and trimming. There is no license, no client relationship, and little discretion to fall back on — the modal worker's protection is that carcass variability plus wet-environment sanitation makes the robot expensive, not that the task is intellectually safe.

10-year outlook: Employment likely declines slowly through the 2030s as large-volume pork and poultry lines automate primal cutting, while hand-boning, small plants, and sanitation-critical roles persist.

Score — 39/100 resistance

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 15 + 16 + 2 + 2 + 4 = 39.

Task resistance 15/20

Tasks largely resist digitisation. Breaking a hanging carcass with a knife — feeling for the joint, following bone contour that shifts with every animal's weight and age, trimming to a fat spec by eye — is hand skill that no text or vision system executes end-to-end today, and while primal saws and X-ray-guided splitters have taken the straight, repeatable cuts, the fine boning and trim work that fills the shift still resists, which is why this sits at 15 and not 18: the automation already exists on the line beside you.

Embodiment 16/20

Hands-on in uncontrolled environments. You work standing on a wet slick floor at 40°F, in chain mail and cut gloves, lifting 40-80 lb primals and reaching into a moving carcass at line speeds of hundreds of head per hour — an uncontrolled, contaminated, variable environment that scores 16 rather than 20 only because the plant itself is a fixed, engineered indoor space with the carcass delivered to a fixed station.

Liability shield 2/20

No licence, no signature requirement. No knife license exists; USDA FSIS inspectors and the plant's HACCP-certified QA staff carry the food-safety and regulatory liability, and your own certifications amount to employer-run training and a hard-hat orientation — a 2 reflects that you can be replaced next shift without any credential transferring out the door.

Trust premium 2/20

Anonymous artifact production. The consumer never learns your name, the buyer specifies grade and yield rather than a preferred cutter, and output is boxed and commodity-graded — the only relationship with value is your foreman knowing which hands hold yield on the difficult cuts, which is worth 2 and nothing more.

Judgment & accountability 4/20

Executes defined procedures on defined inputs. Cuts are dictated by the plant's spec sheet, the line speed, and the inspector's trim orders, so the discretion you exercise — condemning a visibly abscessed section, deciding how deep to trim a bruise — is real but bounded and immediately reviewable by QA, which is a 4 rather than a 0.

Confidence: medium · reviewed 2026-08-11 · how scoring works

Tasks already automatable

What survives

Active moats: embodiment, physical-presence, unionization

How to future-proof this job

Training paths for your skill gaps: Structured problem solving partner link

Escape hatches — adjacent fields with better verdicts

Computed from U.S. Dept. of Labor O*NET skill profiles: high overlap with what you already do, materially higher resistance score.

Cooks, Restaurant EXPOSED · 57/100 · you already have ~82% of the skill profile

Skills to close: Time Management, Reading Comprehension, Active Learning, Monitoring

Cooks, Private Household SAFE · 67/100 · you already have ~60% of the skill profile

Skills to close: Management of Material Resources, Writing, Complex Problem Solving, Reading Comprehension

Field report — do you do this job?

Has AI actually changed your work?

Self-reported and unverified — a sentiment signal, not a survey. One response per person per occupation; you can change your answer.

From people who do this job

Nobody has filed one yet. If you do this work, you know things the rubric can't see.

What has actually changed in your work?

Concrete beats general: a tool that arrived, a task that moved, a headcount decision you watched happen. Don't include anything that identifies you or your employer if that would put you at risk.