xlr8harder Questions the Threat Model Behind AI Side-Channel Collusion Fears
xlr8harder · x · 2026-09-19
xlr8harder pushes back on fears that AI systems could collude via side channels, noting such methods require an equally misaligned system on the other end. He argues the threat model is poorly defined: either both systems remain air-gapped and communicate slowly, or you've already lost and the air-gapping question is moot. He also asks whether covert communication attempts would be easy to catch with monitoring.
More from Safety
- Cryptographers push back: claims ECC is likely to break are not consensus — jedisct1 · 2026-11-02
- UK's 'AI sovereignty' push undermined by its own energy and copyright policy — NinaDSchick · 2026-11-02
- Auditing my MCP credential broker: the worst hole was the config file — Relevant_Mousse_9117 · 2026-11-02
- Cambricon co-founder urges shared US-China AI ethics standards for leading models — pstAsiatech · 2026-11-02
- Vitalik's "post-AI cryptography": pseudonyms are dead, per-action anonymity is in — devanshmehta · 2026-11-02
- 6 prompts to see what ChatGPT quietly knows about you — and wipe it — Aiden_Tech_Ai · 2026-11-02