Double Blind Evals: Resolving the Dual Confidentiality Dilemma in AI Safety...storage.googleapis.com·2 pts·williamtrask·0
Today's Frontier AI companies will never exceed the AI capability frontier againandrewtrask.substack.com·26 pts·williamtrask·9
Frontier AI companies will never exceed the capability frontier againandrewtrask.substack.com·9 pts·williamtrask·0
Zero-Setup Federated Learning: Train Models Across Private Datasets with GColabopenmined.org·1 pts·williamtrask·0
AI First Hiring, Teamwork and Org Structures, Staying Relevant in an an AI Worldmadhavajay.com·6 pts·williamtrask·0
The Cold Start Problem: Using Network Effects to Scale Your Product – A Reviewmadhavajay.com·80 pts·williamtrask·23
Evaluating the State of AI Agent Browser Behavior: How LLMs Access Your Websitemadhavajay.com·1 pts·williamtrask·0
Released Today: US Government Report on Regulating Open/Closed Source AI [pdf]ntia.gov·3 pts·williamtrask·0
Structured Transparency: a framework for addressing use/mis-use trade-offsarxiv.org·2 pts·williamtrask·0
OpenMined open source software 1:1 mentorship program announcementblog.openmined.org·7 pts·williamtrask·0
Investing in privacy enhancing tech to advance transparency in MLblog.twitter.com·1 pts·williamtrask·0