"We will make future versions of 5.X much better at writing" "The future is in general-purpose models"
LLMS
-

Claude Code to get prompt suggestions and new Thinking effort selector
By
–

Besides this – Claude Code is about to get prompt suggestions. – A new Thinking effort selector will be added to the model selector.
-

Moonshot AI begins rolling out Kimi K2.5 on mobile app
By
–


Moonshot AI has begun rolling out Kimi K2.5 on its mobile app. Testing time
-
AI Code Tools Disrupt Developer Pride and Expertise
By
–
It feels like I’m cheating. Which is a very weird feeling to have. It takes a while to unpack. It’s because some code that used to be a point of pride and high IQ and knowledge is suddenly free and instant and it’s very disorienting.
-

Early Grok 4.20 checkpoint achieves +10% gain on Prediction Arena
By
–


An early Grok 4.20 checkpoint has been spotted on Prediction Arena, achieving +10% gain after a 2 weeks long round. Soon?
-
Understanding Comprehension Debt in LLM Development
By
–
Love the word "comprehension debt", haven't encountered it so far, it's very accurate. It's so very tempting to just move on when the LLM one-shotted something that seems to work ok.
-
Yi Tay leads DeepMind’s reasoning breakthrough with on-policy RL
By
–
From pre-training breakthroughs to leading reasoning + RL work inside Google DeepMind’s Gemini org, Yi Tay has had a front-row seat to how modern LLMs evolved from “scale + objectives” into the on-policy RL era powering today’s reasoning leap. Now he's helping assemble DeepMind’s… pic.twitter.com/JRxIhp0kYg
— Latent.Space (@latentspacepod) 26 janvier 2026From pre-training breakthroughs to leading reasoning + RL work inside Google DeepMind’s Gemini org, Yi Tay has had a front-row seat to how modern LLMs evolved from “scale + objectives” into the on-policy RL era powering today’s reasoning leap. Now he's helping assemble DeepMind’s
-

Frontier Model Training Enables More Capable Dangerous Open-Source AI
By
–
These attacks scale with frontier model capabilities. Across both OpenAI and Anthropic model families, training on data from newer frontier models produces more capable—and more dangerous—open-source models.
-
Anthropic introduces interactive apps for Claude using MCP Apps
By
–
Anthropic released interactive apps for Claude. Interactive apps can respond with interactive UI Widgets to enable additional usecases.
— 🚨 AI News | TestingCatalog (@testingcatalog) 26 janvier 2026
"MCP Apps is a new extension to MCP that lets any MCP server deliver an interactive interface within any supporting AI product" https://t.co/AtIc1NMvly pic.twitter.com/ijTcAYhA16Anthropic released interactive apps for Claude. Interactive apps can respond with interactive UI Widgets to enable additional usecases. "MCP Apps is a new extension to MCP that lets any MCP server deliver an interactive interface within any supporting AI product"
-
Question about ChatKit widgets support for GPT Store and Claude
By
–
New it I didn’t find yet if ChatKit widgets are supported. Wondering if it is possible to build one app for both GPT Store and Claude