Usage as a share of theoretical exposure
Weekly work use of generative AI (Bick-Blandin-Deming) divided by the share of the US workforce with at least a tenth of tasks exposed to LLMs (Eloundou and co-authors' "GPTs are GPTs"), derived metric exposure_vs_usage_gap.
Reason through thisThe reading
Roughly half of the exposed workforce uses generative AI weekly; the ratio has climbed steadily since 2024.
9 readings drawn from the derived metric exposure_vs_usage_gap; 10 observations in the evidence.
Derived (exposure_vs_usage_gap) from: FRED (Federal Reserve Economic Data); arXiv abstract pages. Bick, Blandin & Deming, Real-Time Population Survey, retrieved from FRED. arXiv.
Why it matters
The paper's central gap - what the technology could touch versus what people actually use it for weekly. Exposure is a ceiling; usage is the diffusion curve under it.
What cuts against it
The denominator is a 2023 model-based exposure estimate; the numerator counts any weekly use, however shallow; exposure and use are measured on different populations.
How it is measured
- series
fred.us_workers.work_use_weekly_share.q - series
arxiv.us_workforce.llm_exposed_10pct_tasks_share.pt - derived metric
exposure_vs_usage_gap(formula in the semantic layer) - source arXiv abstract pages · default tier 6 · arXiv; abstract quotation
- source FRED (Federal Reserve Economic Data) · default tier 6 · FRED terms of use; RPS series by Bick, Blandin & Deming
- proxy behaviour · cadence quarterly · valve product to adoption
Normal = under 60% of exposed workers use the tools weekly three years in; fast = 80% or more, the gap closed. Between is `emerging`.
Applied to metric:exposure_vs_usage_gap.
Latest readings
Why this status, and its history
Evaluator: 0.4904 (arXiv abstract pages, 2026-06-30) is inside the consistent band (hi=0.6). Auto-reason; band rationale: Normal = under 60% of exposed workers use the tools weekly three years in; fast = 80% or more, the gap closed. Between is `emerging`.
- 2026-09-10unmeasured to consistent with normalconf — → 60 · the evaluator (automatic)
Evaluator: 0.4904 (arXiv abstract pages, 2026-06-30) is inside the consistent band (hi=0.6). Auto-reason; band rationale: Normal = under 60% of exposed workers use the tools weekly three years in; fast = 80% or more, the gap closed. Between is `emerging`.
Why this confidence
Rests on 10 observations from FRED (Federal Reserve Economic Data), arXiv abstract pages; best evidence tier 6, grade B; current.
Confidence is independent of status: 90–95 multiple strong independent sources; 70–89 good evidence, some ambiguity; 50–69 mixed or hard to operationalise; 0–49 limited or vague evidence.
Predictions that lean on this
- Too early to tell