APR 16 · 2026TAKEOVER METER

The AI Takeover Meter · Updated Apr 21, 2026

A single number for how far AI has actually taken over work. Every component is sourced and dated.

Not a vibes check. Not a marketing stat. A weighted composite of labor-market signals you can audit in full below.

Composite score · 0 (quiet) → 100 (total)

42/100

Early disruption — signals are rising but most occupations remain insulated.

Composite of 4 live signals. Last computed 4/21/2026, 8:28:08 AM.

0 — quiet25 — early50 — active75 — late-stage100 — total

Trailing 90 days

The Math

How this number is built.

Each signal is normalized to a 0–100 scale, multiplied by its weight, and summed. Weights are re-normalized so the output stays on the 0–100 scale regardless of which sources are active.

Occupations with declining 10-yr outlook

Share of BLS-tracked occupations projected to shrink over 10 years. Updated annually.

14.00%
raw
46.7
0–100
30%
weight
+14.0
contribution
Source →
Mar 19, 2026

AI model capability (0–100)

A 0–100 composite of frontier-model benchmark scores (MMLU, HumanEval, GPQA).

62.0
raw
62.0
0–100
20%
weight
+12.4
contribution
Source →
Apr 4, 2026

Share of job postings mentioning AI

Percentage of active job postings that mention AI / LLM / machine learning skills.

9.00%
raw
36.0
0–100
25%
weight
+9.0
contribution
Source →
Apr 11, 2026

Announced layoffs attributed to AI (90-day)

Percentage of announced tech-sector layoffs citing AI / automation in the last 90 days.

11.00%
raw
27.5
0–100
25%
weight
+6.9
contribution
Source →
Apr 4, 2026

Missing readings show as "—" and contribute nothing until a fresh reading lands. Weights and sources are managed by the research desk and visible in full here — no hidden fudge factor.

Methodology

Read how we weight and normalize each signal.

Data & Methodology →