AI Deception Warning: People Will Design It to Be Power-Seeking
lxrjl · x · 2026-08-31
Post quotes a dialogue about AI risks, discussing why AI might display deception or power-seeking behaviors. The dialogue suggests that beyond theoretical risks, a direct reason is that 'people will design them to have these traits.' The author notes that with recent actions (HPIM) coming to light, this tweet has aged poorly (i.e., proven too true).
More from Safety
- Scholar asks editors to allow LLM assistance for peer review, gets rejected — RexDouglass · 2026-09-01
- AIF provides pre-execution protection for coding agents — Med1_Ai · 2026-09-01
- Instagram Admits Users Can't Distinguish AI Profiles from Humans — The Decoder · 2026-09-01
- Gary Marcus: AI hypsters bear responsibility for reckless culture — GaryMarcus · 2026-09-01
- Reliance on AI company self-disclosure for incidents is risky — sjgadler · 2026-09-01
- View: Government Attention Is All We Need for AGI Safety — peterwildeford · 2026-09-01