Perplexity and Intelligence
How perplexity measures progress in AI models, and why it could help developers establish a shared approach to evaluating emerging capabilities.
Read the full story on OpenAI News.
How perplexity measures progress in AI models, and why it could help developers establish a shared approach to evaluating emerging capabilities.
Read the full story on OpenAI News.
I asked Google Maps’ Ask Maps, Gemini, and ChatGPT to plan the same Chicago weekend. Ask Maps looked best but got most of the details wrong.
NVIDIA released Alpamayo 2 Edge, a 2B-class reasoning vision-language-action model that plans driving trajectories and explains them on DRIVE AGX Thor, at 160 to 190 ms pixel-to-trajectory latency.
The Binding of Isaac: Repentance Plus has spent the last few days getting review-bombed on Steam after it was discovered that a particular segment of the game used AI music. Reddit user ShweetsStuff discovered the songs used during a Grand Theft Auto-themed section.
The 256GB model is at its all-time lowest price right now.
My discontinued Nvidia card still has a lot of life left in it
Mmathematician at Anthropic says he has used the company’s Claude AI model to prove the Riemann hypothesis, one of the most famous unsolved problems in mathematics. Levent Alpöge made the claim in a note on his website on Friday, Bloomberg reported.
OpenAI’s layered approach to adversarial distillation combines model training, safeguards, detection, and enforcement to protect against capability extraction.
Hebbia tested GPT-6 Astra for financial research. Its prerelease evaluations found gains in source coverage, citation relevance, and presentation quality.
MedChemBench evaluates how well AI models interpret molecular structures, reason from experimental evidence, and select compounds in early drug discovery.
Learn how OpenAI built the Performance Factory, a system of Codex-powered workflows that finds and fixes performance issues to help keep ChatGPT fast.
65 active sources
20 Turkish · 45 English
All sources