Lilian Weng rejoins OpenAI after Briefly Leaving AI Startup
The researcher rejoined OpenAI months after co-founding a rival startup with former OpenAI CTO Mira Murati.
Archive
Category: Artificial Intelligence
Coverage of AI models, machine learning research, generative AI, and artificial intelligence industry developments
Liquid AI releases tiny open-weight encoders for CPU and 8K text
The LFM2.5-Encoder models show that compact bidirectional encoders can still compete on benchmark suites while running on commodity processors.
Sam Altman says AI is unlikely to deliver a shorter work week soon
Altman pushed back against expectations that AI will quickly compress standard work schedules, renewing debate over automation and labor.
Sam Altman says AI will not deliver a shorter work week
OpenAI's CEO says the four-hour-work-week promise was always a mirage, arguing we'll be 'much busier' in a post-superintelligence world.
Perplexity Releases pplx, a JSON-Only Search CLI Built for Coding Agents
The official command-line client offers exactly two commands, deterministic exit codes and token budgeting — and deliberately no chat mode at all.
OpenAI Research: 43.5% of Occupation-Specific ChatGPT Messages Cross Into Other Jobs
An analysis of 800,000-plus messages finds AI is quietly expanding what workers do — a pattern OpenAI calls task crossover, strongest in small businesses and design…
Moonshot AI Releases Kimi K3 Weights: the Largest Open-Weight Model Ever Published
The 2.8-trillion-parameter Mixture-of-Experts model lands on Hugging Face under a custom license — with a 1.4TB download, a $20 million revenue carve-out, and distillation accusations still…
Claude Opus 5 Matches Anthropic’s Priciest Model at Half the Token Cost
Anthropic's new Claude Opus 5 posts top coding and knowledge-work scores while approaching Fable 5's performance at half the token price, with a startling ARC-AGI-3 result.
Genesis AI is raising $500M at a $3B valuation—V Cs bet on robot brains over hardware
The physical-AI startup behind models for robots is in talks for a half-billion-dollar round just a year after its record seed.
Black Forest Labs launches FLUX 3—a multimodal model that generates video with its own audio
FLUX 3 jointly learns images, video and audio, and can produce 20-second clips with native sound—though only in limited release for now.