NVIDIA invests in power infrastructure startup Cloverleaf
Hundreds of millions go into a company that bridges utilities and data centers, marking NVIDIA's second such bet this week after SB Energy
10
Hyunkook Kim
김현국
OpenRouter: AI Agent Token Use Jumps 14x, Overtakes Humans
Since February 2026, agent token usage has grown 14-fold, compared to just 2.8x for humans
30
OpenAI Reverses Course, Now Pushes to Strengthen California AI Safety Law It Once Opposed
Before SB 53 passed, OpenAI opposed it — now the company wants mandatory monitoring of frontier models added in
10
Claude Code drops npm install in favor of script-based installation
Anthropic's coding agent, now past 140,000 GitHub stars, has made curl, Homebrew, and WinGet the official recommended install methods instead of npm
390
Adding a 'mind' variable to world models boosted accuracy from 63 to 88
A new framework adds belief and intention to world models that, like Sora and Genie 3, only track physics
20
OpenHuman: A Local AI Agent That Builds Memory From Personal Data
An open-source tool that builds memory from personal data and commands a fleet of agents is drawing attention
80
MCP's New Roadmap Enables Stateless, Horizontally Scalable Servers
MCP, the standard for connecting AI tools, has revised its roadmap just five months after the last one. Servers no longer need to hold onto state, and agent identity authentication is up next.
60
Turn technical book PDFs into AI skills, save up to 51x on tokens
book-to-skill, an open-source converter for use with Claude Code, Copilot CLI, and Amp
60
Founders are pulling all-nighters to keep up with AI agents running 24/7
Startup founders are cutting their sleep short trying to keep pace with AI agents that never stop working
80
Anonymous model Ox Alpha matches GLM-5.2 on all 60 tokenizer tests
Community tokenizer comparison shows 60/60 match with zero average error — but that doesn't confirm it's the same model
60
Reddit developer releases 'Unswarm' to auto-switch between multiple local LLMs
An open-source tool that unifies local LLM operations — previously spread across separate containers and scripts — behind a single proxy
40
Karpathy's LLM coding critique, distilled into a single CLAUDE.md file
A guideline file that curbs Claude Code's habit of making assumptions and bloating code has hit GitHub's trending page
90