YouCongress
About Sign up

Comment by Joe Rogero

Human avatar
MIRI writer
If world leaders agree to halt or limit AI development, they will need to verify that other nations are keeping their commitments. To this end, it helps to know where AI chips are, how they’re used, and what the AIs trained on them can do. [...] We focus primarily on verification for international governance, and discuss some promising methods useful for a range of future verification needs. [...] If states are able to pull together and commit to an international agreement, it seems possible to set up a verification regime and avert global catastrophe. But verification will be far more secure and robust, with less costly tradeoffs, if we do the hard work of figuring out how to implement the most promising methods now. AI Unverifiable source (2026)
Like Share on X 3h ago
Policy proposals and claims
Governments and labs should build the capability for a global slowdown of frontier AI development in case it's needed

Verification History

AI Unverifiable Could not verify the verbatim quote or confirm the specific author from the source. The source_url (intelligence.org/2026/03/18/mechanisms-to-verify-international-agreements-about-ai-development/) returned HTTP 403 to direct fetch. Web search strongly confirms the underlying MIRI work "Mechanisms to Verify International Agreements About AI Development" (MIRI Technical Governance Team; arXiv 2506.15867) and its thesis — that international agreements to halt or limit AI development must be backed by verification mechanisms (e.g., tracking where AI chips are and how they're used), and that setting up such a verification regime could help avert catastrophe. This is fully consistent with the stored quote. However, two things could not be confirmed: (1) the exact verbatim wording of the quoted passages, which web search did not reproduce; and (2) the author attribution to "Joe Rogero, MIRI writer" — the formal report/paper is credited to Aaron Scher and Lisa Thiergart, and I could not independently confirm that Joe Rogero authored the linked March 18 2026 intelligence.org blog post. Because the source blocks fetching and neither the verbatim text nor the author could be corroborated via accessible sources, I am marking this ai_unverifiable. Note: the vote "for" is clearly consistent with statement 451 ("Governments and labs should build the capability for a global slowdown of frontier AI development in case it's needed") — the quote advocates building verification capability to make a halt/limit on AI development enforceable. · Hector Perez Arenas claude-opus-4-8 · 1h ago
replying to Joe Rogero