Future of Life Institute on Mandate reporting of AI training runs a…

Future of Life Institute

Nonprofit on existential risks

We commend the BIS rule’s inclusion of a provision which requires the reporting [...] computing power greater than 10^{26} integer or floating-point operations.

Disputed source (2024)

7mo ago

Policy proposals and claims

Mandate reporting of AI training runs above 10²⁶ FLOPs to a designated national or international authority

votes For

Statement relation verification history Unverified Report this

No statement relation verification comments yet.

Vote inference verification history Unverified Report this

No vote answer verification comments yet.

Quote authenticity verification history

Report this

Disputed The source page is real and was published on 2024-10-11, but the supplied wording is not a faithful verbatim quote. At the source URL, one passage says the rule "requires the reporting of results from any red-teaming exercise," while a separate footnote says models may involve "computing power greater than 10^{26} integer or floating-point operations." Those are separate passages, not one continuous quote, and splicing them with [...] materially changes the meaning. The page also lists Hamza Chaudhry as the webpage author and Future of Life Institute as the organization/submitter. · YouCongress gpt-5.4-2026-03-05 · 3mo ago

Disputed The source URL is a Future of Life Institute document published on 11 October 2024; it lists Hamza Chaudhry as the author and Future of Life Institute as the organization. ([futureoflife.org](https://futureoflife.org/document/bis-rule-for-establishment-of-reporting-requirements/)) But the submitted quote is not verbatim: the first fragment appears in section 2 about reporting red-team results, while the 10^{26} wording appears separately in footnote [1] about model scope. ([futureoflife.org](https://futureoflife.org/document/bis-rule-for-establishment-of-reporting-requirements/)) The ellipsis therefore stitches together different, non-contiguous passages and materially alters the original wording/context. ([futureoflife.org](https://futureoflife.org/document/bis-rule-for-establishment-of-reporting-requirements/)) · YouCongress gpt-5.4-2026-03-05 · 3mo ago

replying to Future of Life Institute

Other opinions from Future of Life Institute

See all

Future of Life Institute votes For

Open-source AI is more dangerous than closed-source AI

Releasing full model weights may allow malicious actors to strip or override safety mechanisms, creating uncensored or harmful versions. In contrast, supervised fine-tuning preserves core safety guardrails while enabling responsible customization.

AI Verified source (Jul 2026)

added 1mo ago

Future of Life Institute votes For

Require AI systems above a capability threshold to be interpretable

Leading AI companies should: Ensure Human-Legible Reasoning - AI models should reason in ways that are accessible and understandable to humans. Developers should avoid opaque reasoning methods.

AI Verified source (Jul 2026)

added 2mo ago

Future of Life Institute votes For

Grant whistleblower protections to AI safety researchers