The Compute Verification Post
substack.com · 5,253 words · saved by 1 readers
Notes from our nights and weekends
Summary: We start with a primer on compute verification: understanding what’s going on inside a datacenter and how we can prove it to someone else. This involves proving certain properties about compute, e.g. where it is, what it’s doing, and how much was used, under particular trust assumptions. We then argue: Inference-only verification is the highest-value near-term property and the highest-priority target for large-scale demo buildout, International verification will never be trustless, only low-trust, so we should focus on stacking decorrelated defense-in-depth measures, Verification…
saved by
related reading
- The Compute Verification Postfirstscattering.com
- Verification Planai-2040.com
- Asymmetry of verification and verifier’s rule - Jason Weijasonwei.net
- [2606.05433] Zero knowledge verification for frontier AI training is possiblearxiv.org
- We Should Think More Critically About Compute Verificationsudarsh.com
- When AI Writes the World's Software, Who Verifies It? — Leonardo de Mouraleodemoura.github.io
- A shallow dive into formal verificationvitalik.eth.limo
- Verifying International Agreements on AI: Six Layers of Verification for Rules on Large-Scale AI Development and Deploymentarxiv.org
- [2004.07213] Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claimsarxiv.org
- STARK-paper.pdfstarkware.co
- The importance of full-stack openness and verifiabilityvitalik.eth.limo
- Private Cloud Compute: A new frontier for AI privacy in the cloud - Apple Security Researchsecurity.apple.com