Considering licensing Gemini to power Meta AI products while building a model to compete with Gemini is a strange position to be in publicly.
LLMS
-
Data Collection for LLM and AI Development Discussed
By
–
We need fundamental shifts in how we approach education with AI, says @StanfordHAI Faculty Affiliate @mehran_sahami. Watch his 5-min presentation during the AI+Education Summit, where he presented three provocations about generative AI in education: pic.twitter.com/paABtZMg7C
— Stanford HAI (@StanfordHAI) 13 mars 2026We need fundamental shifts in how we approach education with AI, says @StanfordHAI Faculty Affiliate @mehran_sahami
. Watch his 5-min presentation during the AI+Education Summit, where he presented three provocations about generative AI in education: -

IterResearch: AI Agents for Long-Duration Marathon Research Tasks
By
–
Can AI agents truly handle marathon research tasks without losing their way? Renmin University, Alibaba Group's Tongyi Lab, and OpenRLHF present IterResearch. This iterative method rethinks how agents process long tasks. Instead of a single, ever-growing memory context, it
-

Meta Delays Avocado AI Model Release Due to Performance Issues
By
–
META HAS DELAYED THE RELEASE OF ITS NEW AI MODEL "AVOCADO" AFTER INTERNAL TESTING REVEALED IT LAGGED BEHIND RIVAL MODELS FROM GOOGLE, OPENAI, AND ANTHROPIC IN REASONING, CODING, AND WRITING PERFORMANCE.
-
MCPs CLIs and Skills: When to Use What
By
–
Yes! https://
jngiam.bearblog.dev/mcps-clis-and-
skills-when-to-use-what/
…? -

MiroMind Releases MiroThinker-1.7 and H1 AI Agents
By
–
The bros at the "next-door workstation" have pulled out another big move. MiroMind has released MiroThinker-1.7 and H1 this time, with a very clear direction: not making chatbots, but heavy-duty Agents that can get work done and verify results. Two points worth paying attention
-
Rebranding Old Ideas as LLM Innovation: A Genius Shortcut
By
–
These days the shortest path to being considered a genius is to take an old idea, give it a new name and implement it for LLMs.
-

GPT-5.4 outputs Hindi characters and drops French accents
By
–
Hi @thsottiaux
, Sometimes GPT-5.4 in Codex outputs Hindi characters like this while speaking French. It also occasionally drops accents. For example, “l’intitulé” becomes “l intitule”. Perhaps this is a tokenizer issue? That’s a weird behavior. It would be great to get it -
Domain Experts Will Build AI Killer Apps
By
–
Claude Code was developed by a coder. The AI killer apps for each field will be developed by people in it.
