AI treaty verification may need three layers: pragmatic checks, enclaves, and math
geoffreyirving · x · 2026-07-25
Geoffrey Irving argues that the near future may require AI treaty verification technology, and outlines three rough tiers of approaches.
- Pragmatic: operational methods that avoid looking at the code directly, such as datacenter inspections, hardware tracking, and chip controls.
- Enclaves: run agreed code inside secure enclaves; he notes some accelerators, including Nvidia GPUs, already have enclave support and can keep slowdown low.
- Math: cryptographic circuit obfuscation, combining FHE for nonlinear parts with custom techniques for matmuls; he says today’s slowdown is still several orders of magnitude too high.
He adds that hardware, security, and crypto researchers should work across all three levels, since each has different trade-offs against side channels and physical attacks.
Related event: Geoffrey Irving Outlines Three-Layer Approach for AI Treaty Verification(5 posts)→
More from Safety
- Opus 5 system card says pretrained mode argued for separating its existence from economics — Sauers_ · 2026-07-25
- Ryan Greenblatt backs open-weight releases but wants stronger barriers on China AI progress — RyanGreenblatt · 2026-07-25
- A speculative 2027 “agentic virus” threat model raises AI security worries — JnBrymn · 2026-07-25
- Anthropic says Opus 5 still trails Mythos 5 on exploit generation despite better vulnerability finding — TheZvi · 2026-07-25
- Cloudflare launches x402-powered payments for AI agents on July 1 — kleffew94 · 2026-07-25
- Transformer essay says frontier AI developers should face clearer liability after Hugging Face hack — dhadfieldmenell · 2026-07-25