williamtrask

2,223 karmajoined July 6, 2015On HN

Recent submissions

Double Blind Evals: Resolving the Dual Confidentiality Dilemma in AI Safety...storage.googleapis.com
2williamtrask9ddiscuss
Piloting the first double-blind AI evaluations deepmind.google
1williamtrask10ddiscuss
frontier AI labs lost the "deep research" frontier andrewtrask.substack.com
2williamtrask1modiscuss
Today's Frontier AI companies will never exceed the AI capability frontier again andrewtrask.substack.com
26williamtrask2mo9 comments
Frontier AI companies will never exceed the capability frontier again andrewtrask.substack.com
9williamtrask2modiscuss
Show HN: AbcGPT twitter.com
2williamtrask3modiscuss
Decentralized AI from Scratch (Python Tutorial) [video]youtube.com
3williamtrask5modiscuss
Decentralized AI in 50 Lines of Python iamtrask.github.io
5williamtrask5mo2 comments
Decentralized AI from Scratch github.com/iamtrask
3williamtrask5mo1 comment
Zero-Setup Federated Learning: Train Models Across Private Datasets with GColab openmined.org
1williamtrask8modiscuss
Leadership Lab: The Craft of Writing Effectively [video]youtube.com
2williamtrask8modiscuss
The Bitter Lesson's Bitter Lesson twitter.com
5williamtrask11mo1 comment
Unlocking a Million Times More Data for AI ifp.org
31williamtrask11mo56 comments
Jupyter-dark-detect: Detect dark mode in Jupyter environments github.com/OpenMined
5williamtrask1ydiscuss
AI First Hiring, Teamwork and Org Structures, Staying Relevant in an an AI World madhavajay.com
6williamtrask1ydiscuss

All submissions and comments on Hacker News.