How We Calculate Risk
Every score is reproducible. No black boxes. Two methodologies power Takeover Tracker: the daily macro index and per-occupation risk scores.
Daily Displacement Index
The macro-level AI takeover score updated daily
Collect
9 async collectors pull signals from RSS, BLS, FRED, GDELT, Reddit, HN, SEC EDGAR, layoff trackers, and benchmarks.
Extract
Gemini classifies each signal: category, sectors, sentiment, impact score, and displacement direction.
Score
24 parallel scoring calls (4 sectors x 6 categories) with self-consistency sampling (3x median).
Calculate
Pure math: weighted geometric mean, hype discount, dual-EMA smoothing, 2-point daily movement cap.
Signal Category Weights
How each data source type contributes to the daily score
BLS employment, job postings, layoff data — most reliable ground truth
Earnings calls, enterprise AI spending, hiring patterns
Model benchmarks, task performance — hype-discounted
GDP per worker, productivity data, sector output
Social media, news volume — heavily discounted
AI legislation, executive orders, industry standards
Job Sector Weights
30%
Knowledge
Legal, finance, consulting
25%
Service
Support, sales, admin
25%
Technical
Software, data, IT
20%
Creative
Writing, design, marketing
Scoring Formula
Geometric Mean Aggregation
HDI method — prevents one hot sector from inflating the overall index. Each sector's weighted score contributes proportionally.
Hype Discount
When hype signals exceed 2x reality signals, capability scores are reduced 15% and sentiment 50%. Keeps the index grounded.
Dual-EMA Smoothing
70% slow (~90-day) + 30% fast (~20-day) exponential moving average. Prevents day-to-day noise from dominating.
Movement Cap
Maximum 2-point daily change with 30% carry-forward. Large swings are dampened and partially applied over subsequent days.
Source Credibility Hierarchy
Scoring Thresholds
Each score level maps to observable labor market conditions
Per-Occupation Risk Scoring
Individual displacement scores for 900+ occupations using O*NET task data
Task Risk Formula
Every task in an occupation is classified into one of five categories and scored for AI capability:
Tasks are weighted by estimated time fraction to produce the raw score:
Task Categories & Base Risk
Higher base risk = more automatable task type
Protective Factors
Five factors can reduce the raw score by up to 55%:
Empathy, negotiation, reading social cues
Novel ideation, artistic expression, innovation
Ambiguous judgment, ethical reasoning, strategic calls
Licensing, legal requirements, safety standards
Dexterous physical tasks, precision work
Risk Tiers
75 – 100
Critical Risk
50 – 74
High Risk
25 – 49
Medium Risk
0 – 24
Low Risk
Explore the data yourself
Search 900+ occupations, compare risk scores, and see exactly how each job is scored.