EXPOSURE TO AI
16%
RESILIENT
OBSERVED IN REAL USE · Anthropic 2026
2%
of this role’s work is already showing up in real Claude usage (Anthropic Economic Index).
Robots can’t hold a hand, a scalpel, or your nerve.
16% of this role’s O*NET tasks are within reach of today’s AI. That is the core-weighted exposure score from Eloundou et al. 2023 (“GPTs are GPTs”). It measures a capability ceiling, not a headcount forecast. Certified hard to automate. Today’s AI barely touches the core of this one.
WHAT AI CAN ALREADY DO
- Record information about work completed and machine settings
- Program electronic equipment
- Inspect products to ensure that the quality standards and specifications are met
- Study guides, samples, charts, and specification sheets or confer with supervisors or engineering staff to determine set-up requirements
WHAT IT STILL CAN’T
- Place patterns on top of layers of fabric and cut fabric following patterns, using electric or manual knives, cutters, or computer numerically controlled cutting devices
- Start machines, monitor operations, and make adjustments as needed
- Adjust machine controls, such as heating mechanisms, tensions, or speeds, to produce specified products
- Inspect machinery to determine whether repairs are needed
- Repair or replace worn or defective parts or components, using hand tools
THE HONEST PART. A percentage is not a pink slip. High exposure usually means a role shrinks and shifts toward judgment, direction and responsibility: the parts a model can’t sign its name to. Exposure ≠ displacement. Breathe.
"My job is 16% cooked. What’s yours?"
SOURCES: O*NET 30.3 occupational tasks · Eloundou et al. 2023 (“GPTs are GPTs”,
arXiv:2303.10130) · Anthropic Economic Index 2026 (CC-BY) |
how this is calculated |
last updated 2026-07-16