Loading activity…
@andrewdai
Co-founder and CEO at @ElorianAI. Ex (director) at Google DeepMind after 12y. Gemini Data Area Lead. Now building at the visual frontier. 戴明博
AI
Glimpse’s team claims competitive frontier performance compared with Astra on key public visual reasoning benchmarks.
AI
A panelist says AWS Startups is hosting the October 14 session, with Reactor's CTO also on the panel.
AI
Scale Labs says humans scored 93.1% across 522 tasks, versus 53.6% for top model GPT-6 Astra.
AI
Citing MechReason, a post says models fail most at connecting the different kinds of evidence engineers use to diagnose failures.
AI
An ElorianAI team member says benchmarks show models recognizing patterns in image sequences but repeatedly choosing the wrong image to complete them.
AI
Amodei says Anthropic will give third-party evaluators permanent, employee-level access to its systems to verify adherence to safety measures, report incidents and assess models’ alignment during training.
AI
The host’s agenda asks how automated AI researchers will be trained, whether long-horizon reinforcement learning will elicit artificial general intelligence, and how much progress is explained by data.
AI
“Wife told me these wouldn’t fit,” one user wrote, joking: “Little did she know I had trained for this moment for years.”
AI
AI's broader abilities take work to draw out, one user argues. They suggest building a 3D environment and using toy problems to develop intuition alongside a model.
AI
François Chollet and ARC Prize discuss GPT-6 Astra's benchmark results and what they mean for AGI claims.
AI
Elorian AI CEO discusses model struggles with video and image comprehension.
AI
Axiom Math researchers announce a human-AI result that lowers the bound on prime gaps to 212.
AI
Four former Google researchers found a public benefit corporation to automate machine learning and science.
