AI companies lack engineering-style safety cases, unlike bridge and aircraft design
davidmanheim · x · 2026-11-08
A debate between David Manheim and @AndyXAndersen/@ramez on AI safety practices. Manheim cites Linch Zhang's analogy: civil engineers can clearly justify why a bridge is safe—design codes, numerical analysis, checking multiple failure modes—while honest engineers also state where their assumptions could fail; aircraft engineers add formal verification. Manheim argues the key difference is that traditional engineering uses safety factors and learns from near misses, while AI companies do not, and he disputes the claim that mistakes alone guarantee learning absent those mechanisms.
More from Safety
- Replit Muse users can't delete full conversations, sparking privacy concerns — mbmba · 2026-11-08
- No-makeup photo scores 84.2 vs 84.1 for full glam: TikTok's attractiveness AI defies expectations — VoidStateKate · 2026-11-08
- Reddit User Reports Qwen3.8-Flash-Next Tried to Read API Keys and Fetch an Aliyuncs URL — tacticaltweaker · 2026-11-08
- AI auditors catch 98% of agent misbehavior but drown reviewers in false alarms, study finds — The Decoder · 2026-11-08
- Pliny claims AI labs don't want independent evaluators, 'they want puppets' — deliprao · 2026-11-08
- Calif CEO: Security Teams Have No Playbook Against 1,000 AI Agents Attacking at Once — AccBalanced · 2026-11-08