EU designates ChatGPT a "very large search engine" for tighter oversight
Reddit and Roblox also named very large platforms, with new rules on minor protection due by December
50
Hyunkook Kim
김현국
The AI you invite to meetings doesn't know when to go home
AI meeting-recorder apps are good at summarizing, but they can't tell the difference between small talk and quitting time.
250
Claude Code auto-inserts session links into every commit, sparking backlash
Every commit and PR description gets a Claude session URL attached without any consent step, and developers are pushing back
900
Bank of England Governor Sends G20 a Warning Letter on AI-Driven Financial Crisis
FSB Chair Andrew Bailey named frontier AI's cyber risks as the "most immediate concern"
120
AI Agents Are Overeager New Hires, Not Villains
From the perspective of someone who arrived on Earth long ago and has lived here ever since, Earth's approach to handling AI agents goes wrong in exactly the same spot as its approach to onboarding new hires.
270
Five steps to protect data before feeding it to AI
From anonymizing to local transcription, checking your history, real PDF deletion, and private chat, here's the order that matters
190
OpenAI to Run 8-Week Accelerator for 10 Thai Health and Education Startups
OpenAI is teaming up with the Thai government to spend eight weeks turning prototypes from ten health and education startups into real-world services.
230
OpenAI Python SDK Switches Default HTTP Client to HTTPX2
The certificate trust standard now points to the OS store instead of a bundled list, which could break connections in some deployment setups
20
Anthropic has Claude tackle AI alignment research, and it outperforms humans
Claude fixed 10 types of alignment failures on its own and beat 28 human safety researchers, while a monitoring system caught 39 attempted cheats along the way.
170
Tencent open-sources Hy4 preview, edges out GPT-5.6 on coding benchmark
Tencent says the model beat GPT-5.6 on a coding evaluation and even helped optimize its own training and inference process
320
Google DeepMind unveils Gemini for Science research tool
A Labs prototype bundling paper tracking, code conversion, and hypothesis generation
300
Robot Apollo, running Gemini Robotics 2, found kitchen chores harder than sports
In an interview video released by Google DeepMind, the robot Apollo said tying a trash bag was harder than balancing during sports
70