Skip to content

jschulman/displacement-curve

Repository files navigation

The Displacement Curve

Tracking AI's displacement of human labor in professional services — with data, not opinion.

View the live dashboard →

What Is This?

An open-source dashboard that aggregates publicly available market signals to answer one question: Where are we on the displacement curve?

The Composite Displacement Index synthesizes eight independent signals into a single 0-100 score:

Score Phase What It Means
0–25 Pre-disruption AI is a topic, not a force
26–50 Productivity AI making firms more efficient; employment stable
51–75 Erosion Revenue per employee diverging from headcount
76–100 Displacement Employment declining; funding pouring into replacements

Signals

Signal Source Frequency What It Measures
Professional Services Employment BLS (CES) Monthly Sector headcount trends
AI Search Interest Google Trends Daily Public attention to AI tools
Open Source AI Activity GitHub API Daily Developer momentum in AI
AI Revenue Reporting SEC EDGAR (XBRL) Quarterly How firms report AI revenue
Revenue Per Employee SEC EDGAR (10-K) Quarterly Productivity divergence signal
VC Funding: AI Services SEC Form D Quarterly Capital flowing to AI replacements
AI vs Traditional Hiring BLS JOLTS Monthly Job opening mix shift
Regulatory Guidance Fed / OCC / SEC / NIST / EU Quarterly Government response velocity

Every signal uses free, public data sources — no proprietary APIs, no paywalls.

Design Principles

  • Data, not opinion. The dashboard presents signals. It does not editorialize.
  • Reproducible. Every data point traces to its source. Every collector is open source.
  • Automated. GitHub Actions collects fresh data on schedule.
  • Durable. Only signals with reliable, ongoing public sources are included.
  • Accessible. Single page. No login. No paywall. Mobile-friendly.

Architecture

collectors/          Python scripts for each data source
normalizers/         Derived metrics (composite index, earnings normalization)
data/                Versioned JSON (raw + processed)
docs/                Static dashboard (HTML/JS/CSS — GitHub Pages)
.github/workflows/   Automated collection schedules
Layer Technology
Dashboard Vanilla HTML/JS/CSS
Charts Chart.js (CDN)
Collectors Python 3 + requests
Scheduling GitHub Actions (cron)
Hosting GitHub Pages

Running Locally

# Install dependencies
pip install -r requirements.txt

# Generate mock data for development
python3 data/generate_mock_data.py

# Or run a live collector (example: BLS employment)
python3 collectors/bls_employment.py --api-key YOUR_BLS_KEY

# Serve the dashboard
cd docs && python3 -m http.server 8000

All collectors support --mock for offline development.

Methodology

See METHODOLOGY.md for detailed documentation of data sources, collection methods, normalization, and composite index weights.

Contributing

See CONTRIBUTING.md for guidelines.

Short version: fork, branch, follow existing patterns, test locally, submit a PR.

License

MIT — see LICENSE

About

Tracking AI's displacement of human labor in professional services — with data, not opinion.

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages