EXPOSURE TO AI
44%
MODERATE
OBSERVED IN REAL USE · Anthropic 2026
12%
of this role’s work is already showing up in real Claude usage (Anthropic Economic Index).
Half in the fire, half out. Choose which half you become.
44% 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. In the blast radius. A real slice of the work is already automatable. The rest isn’t.
WHAT AI CAN ALREADY DO
- Prepare correspondence, reports of inspections or investigations, or recommendations for action
- Examine records, reports, or other documents to establish facts or detect discrepancies
- Inspect government property, such as construction sites or public housing, to ensure compliance with contract specifications or legal requirements
- Investigate alleged license or permit violations
- Collect, identify, evaluate, or preserve case evidence
WHAT IT STILL CAN’T
- Submit samples of products to government laboratories for testing, as required
- Inspect government-owned equipment or materials in the possession of private contractors to ensure compliance with contracts or regulations or to prevent misuse
- Testify in court or at administrative proceedings concerning investigation findings
- Coordinate with or assist law enforcement agencies in matters of mutual concern
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 44% 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