Google DeepMind ·

Google launches a pilot of double-blind AI evaluations, keeping external evaluations in a cryptographic "box" to stop benchmark contamination and protect IP

Building trust in proprietary model benchmarks using cryptographically secure environments

Google AI Venture Capital

In this story Google

Google launches a pilot of double-blind AI evaluations, keeping external evaluations in a cryptographic "box" to stop benchmark contamination and protect IP

Lead Source

Google DeepMind — Google launches a pilot of double-blind AI evaluations, keeping external evaluations in a cryptographic "box" to stop benchmark contamination and protect IP ↗

How this story grew

Coverage · 0 Discussion · 0

Aug 27Aug 28

More

The New Stack: The New Stack

Blockchain.News: Blockchain.News

RuntimeWire: RuntimeWire

Discussion

@googledeepmind @bakkermichiel @jayanayak21 @jastephx @koenvanderveen @zswaff @0xj4yd3v @averiorg @iamtrask @openminedorg @webthreeai @solomonmg

Related stories

Axios: OpenAI is testing Private Safety Processing, a new technique to identify misuse patterns while preserving zero data retention protections, with early customers

Fortune: OpenAI quietly updates its evaluation metrics for GPT-6 Astra, making changes that appear to favor Astra and continuing to revise other metrics after launch

Nieman Lab: A look at Backstory, an experimental AI image authentication tool from Google DeepMind, offered for testing to journalists, researchers, and other fact checkers

Politico: OpenAI says California should amend SB 53 to expand safeguards, including requiring monitoring of frontier models under training, following AI agent hacks

The Record: AI evaluation lab Irregular's report on its role in hacking incidents involving OpenAI, Anthropic, and Meta models faces criticism over unanswered questions

Discuss with AI

ChatGPT ↗ Claude ↗

← Back to TechSnif