Projected 2027 misaligned AI damages: $624m vs $617bn from cars
joshua_saxe · x · 2026-10-11
AI safety researcher Joshua Saxe projects 2027 damages from misaligned agents at $624m under current trends—dwarfed by cars ($617bn), medical errors ($127bn), and manual IT errors ($246bn). He argues AI will inevitably cause damage as economic substrate, and the real questions are pareto-efficient damage-benefit regimes, liability allocation, and whether AI firms get regulated like gilded age railroads—while still taking catastrophic black-swan risk seriously.
Related event: Study Estimates Misaligned AI Agents Will Cost $624M in 2027(2 posts)→
More from AGI Musings
- The 'freeswarm' thesis: autonomous agent swarms that fund themselves could arrive by 2027 — max_paperclips · 2026-10-11
- Rick Lamers: Don't normalize cruelty in AI conversations just because it's 'Normal Technology' — ricklamers · 2026-10-11
- Yoav Goldberg blasts COLM mood: RL loss tweaks aren't 'science again' — yoavgo · 2026-10-11
- Ex-US AI Envoy Rumman Chowdhury Pushes Back on AI Doom Narratives — math_rachel · 2026-10-11
- Essay sparks debate: AI has no stakes in the living world yet will decide its meanings — begusgasper · 2026-10-11
- "If you get good, you can literally print money": AI era rewards skilled labor — teodorio · 2026-10-11