AI safety debate: incident reporting and compute-threshold audits as low-hanging fruit
BlackHC · x · 2026-09-11
In an exchange on AI safety regulation, participants agreed that slogans like 'regulate more' or 'ban everything' are unproductive without concrete, workable proposals:
- BlackHC highlights two low-hanging policy fruits: mandatory incident reporting requirements and audits above certain compute/cap thresholds.
- He notes concrete proposals already exist but lack momentum, urging well-connected people to push them forward.
- sytelus agrees but stresses the need to identify genuinely new, unknown safety issues and think deeply about proposals with a real chance of working.
Related event: AI safety debate converges on concrete, actionable policy proposals(2 posts)→
More from Safety
- 6TB leak from a Chinese LLM router exposes credentials to hijack Xiaomi, Huawei and gov entities — teortaxesTex · 2026-09-11
- Epoch AI data: China's top models trail US frontier by 6.3 months, real gap closer to 8 — deanwball · 2026-09-11
- Biorisk defense is man-made, not natural difficulty: the open-model bioweapon debate continues — JMannhart · 2026-09-11
- Open-source models by 2029 could make bioweapon creation "almost trivial", debaters clash on biorisk — JMannhart · 2026-09-11
- DeepSeek and Moonshot accused of secretly relaying user prompts to Claude via fake accounts — burny_tech · 2026-09-11
- Leo Laporte: AI-created worms are malware, not doom — prosecute the creators — GaryMarcus · 2026-09-11