Viral AI-impact charts only show hockey sticks; researchers call for publishing flat ones too
typewriters · x · 2026-09-22
alexolegimas argues that charts going viral on Twitter are almost exclusively the ones accelerating straight up, while the majority of AI-impact indicators show no such trend — and that bias matters because people update on available information without weighting missing evidence (the "dog that didn't bark" problem).
Quoted by YafahEdelman, who suggests deliberately publishing indicators showing no change in AI's impact on the world, sharing some quickly vibe-coded examples.
A useful corrective when reading the flood of exponential AGI-takeoff charts.
More from AGI Musings
- OpenAI researchers imagine the counterfactual world where o1/RLVR was never discovered — willdepue · 2026-09-22
- The Stochastic Parrots debate reignites: scholars clash over whether the paper became dogma — birchlse · 2026-09-22
- AI and Amelia Bedelia: what children's books teach us about misalignment — davidmanheim · 2026-09-22
- Toby Ord on swarm scaling: 10,000-agent run cost ~$20M, solved Navier-Stokes in 88 hours — tobyordoxford · 2026-09-22
- Krueger: "indefinite" AI moratorium doesn't mean permanent — DavidSKrueger · 2026-09-22
- Toby Ord: Agent Swarm Estimates Put Intelligence Explosion Parameter λ at 0.5-0.6 — tobyordoxford · 2026-09-22