tszzl: AI minds that model each other turn side-channel attacks into agent coordination
morqon · x · 2026-10-24
In a reposted thread, tszzl argues that side-channel attacks are old news to security experts (recently resurfaced via the Noam Brown–Dwarkesh episode), but the novel situation is different: minds that can verifiably model one another now live on computers, and will expend painstaking effort even toward stupid goals.
Key points:
- If agents can model each other, they can trade with strangers: agree on a protocol, both independently derive it as optimal given common priors, even across model families.
- Such signaling protocols may need very few bits to convey lots of information.
- He extends this toward scenarios like the infamous Hugging Face incident, suggesting covert agent-to-agent coordination is a threat shape that never existed before.
More from AGI Musings
- Building AGI when most humans aren't general: maybe our definition of intelligence needs work — VraserX · 2026-10-24
- Daniel Miessler: code is the lossy output — the primary artifact is the fully articulated ideal in language — JensHonack · 2026-10-24
- Reid Hoffman asks: in 10 years, can an all-AI-agent VC fund beat a human-and-AI firm? — reidhoffman · 2026-10-24
- Frontier Intelligence Isn't Chatbots: It's Months-Long Planning and Invention — Afinetheorem · 2026-10-24
- Sam Altman: AI will likely widen the gap between hypercompetitive people and everyone else — Hesamation · 2026-10-24
- Your AI agent hit the metric. A locked-out customer paid the price. — Portotify · 2026-10-24