CopeCheck

The AI Cope Index — tracking what public figures really think about AI replacing jobs, scored by an LLM oracle.

Live at copecheck.com

What it does

CopeCheck monitors 19 high-profile figures (AI lab CEOs, tech leaders, economists, public intellectuals) and scores their public statements on a cope-to-doomer spectrum using an LLM-powered analysis pipeline.

Each figure gets a Cope Score (0–10) based on recent quotes about AI and employment, with higher scores indicating more optimistic/dismissive takes on AI disruption.

Stack

• Backend: Python/Flask
• Database: SQLite
• LLM Oracle: Claude (via Anthropic API) for quote analysis and scoring
• Image Generation: Straico API (DALL-E 3) for editorial portraits
• Hosting: Hetzner VPS behind Cloudflare tunnel
• Domain: copecheck.com

Architecture

• app.py — Flask web app, serves the index and figure pages
• pipeline.py — Automated pipeline: search → collect → analyse → score
• oracle.py — LLM-powered quote analysis and cope scoring
• sources.py — Web search and quote extraction
• db.py — SQLite database layer
• figures.yaml — Configuration for tracked figures and search queries
• seed_quotes.py — Initial data seeding

License

MIT