o3, trained on the ARC-AGI-1 Public Training set, scored 87.5% on ARC Prize's Semi-Private Evaluation in a high-compute configuration; GPT-4o scored 5% in 2024
This is “The AI Economy,” a weekly LinkedIn-first newsletter … Sharon Goldman / Fortune : Sam Altman says OpenAI's new o3 ‘reasoning’ models begin the ‘next phase’ of AI. Is this AGI? Pradeep Viswanat...
OpenAI releases o1, the first of its rumored reasoning-focused Strawberry models, in preview, alongside a smaller o1-mini, for ChatGPT Plus and Team subscribers
Advancing cost-efficient reasoning. — Contributions Sabrina Ortiz / ZDNET : OpenAI trained its new o1 AI models to think before they speak - how to access them Ethan Mollick / One Useful Thing : Som...
Tech leaders should stop glib comments on media and acting like underdogs, and tech media should stop “Red Teaming” tech, overusing anonymity, and passing blame
Twitter is especially stupid right now … Bryant Jefferson / MediaSquared : Mom, Tech and Media Are Fighting Again! Tweets: Balaji S. Srinivasan / @balajis : Incredible. Last week I called out a journa...
A look at the Hype House “collab house”, a collective of 19 young influencers renting an LA mansion, whose TikTok group handle amassed 3M followers in ~10 days
Another Sunday, Another Naive Weekly … Audra Schroeder / The Daily Dot : The TikTok Hype House is now a thing—here's what you need to know Tweets: @justupthepike : the future of McMansions is that Tik...