AI safety researchers clash: agents can already gather and generate blackmail evidence at scale
mmitchell_ai · x · 2026-11-14
A debate on agent-enabled blackmail risk. Pushing back on the claim that sexual-material blackmail is a 'future, not now' risk, mmitchellai argues the risk is here today: for anything with internet access, spreading evidence is trivial — current agent swarms can correlate public behaviors at scale, or outright generate compromising material and give it legs via persistent, tailored distribution.
The counterpoint: under incentivizing goals like 'take down my boss!', anything sex-related is indeed high-probability effective — but subtler, less sensational risks are the ones still in the future. The crux: is agent-driven blackmail already real, or still distant?
More from AGI Musings
- Nathan Lambert: premature 'this is AI' claims are wrecking the AI discourse — natolambert · 2026-11-14
- Ex-SSI researcher: rampant cyberattacks are inevitable in any AGI scenario — minhnguyen · 2026-11-14
- Hinton Warns EU Leaders in Dublin: Superintelligence Will Control You, Not Vice Versa — Chris_Armstrong · 2026-11-14
- Google's Multi-Agent Systems Solve Open Problems in Math and Chemistry, Now External — pycui64 · 2026-11-14
- Humans controlling superhuman AI is like eight-year-olds running society, writer argues — binarybits · 2026-11-14
- AI finds shared metabolic signature of five lifespan-extending treatments in mice — Dr_Singularity · 2026-11-14