AI BriefingDaily AI briefing
Editor's picks come first, followed by stories repeatedly covered across days and sources.
Top stories
05- 01Google DeepMind Launches Gemini 4 Hydrogen for Efficient Coding and Agentic Workflows
Priced at $1.50 per million input tokens, the model targets speed and efficiency for daily coding tasks, distinguishing it from the frontier-focused Gemini 4 Carbon and Argon.
Google AI Blog
- 02Forensic Analysis Finds Blackwell GPU Fingerprints in Six Chinese Frontier AI Models
A forensic report identifies NVIDIA Blackwell hardware signatures in the initial weights of six Chinese models, including DeepSeek-V4-Flash and Kimi K3, indicating access to export-controlled chips.
Hacker News
- 03Row-Bot 5.3.3 adds browser-based deployment, 12 new API providers, and Claude Code context optimization
The update allows running Row-Bot in the browser via Railway, adds 12 new API providers, and optimizes Claude Code to use prompt caching for full 1M context on Opus, Sonnet, and Haiku 5.5.
Reddit
- 04Amplitude's Wave agent uses OpenAI models to ship product improvements in 1–2 days
Amplitude's Wave agent drove a 160% lift in card-driven sends on its homepage by personalizing suggestion cards using session data.
OpenAI Blog
- 05Google A2UI 1.0, Express Method Cuts Input Tokens by 45.4%
Google released A2UI 1.0, reducing input tokens by 45.4% when generating agent UIs.
PyTorchKR
Developer tool releases
01Rising repositories and models
02- kuafuai/andee
A Personal AI Agent That Lives on Your Phone and Mac Without the Cloud
- vercel-labs/ui
A UI Behavior Library Written Directly by Agents