AI Coding
Gemini Omni Flash Turns Video Editing Into a Conversation
Gemini Omni Flash reframes AI video generation as an interactive editing loop connected to Google Flow and YouTube Shorts.
AI Coding
Android Inside AI Studio, and the App Bottleneck Moves to the Browser
Google AI Studio now reaches Android app generation and Play testing paths, moving the AI app bottleneck from prompting to verification.
AI Agents
WebMCP Origin Trial turns browser agents from click guessers into tool users
Chrome WebMCP is an Origin Trial for exposing structured tools and permission boundaries from web apps to browser agents.
AI Coding
AlphaEvolve cuts errors 30% and turns coding agents into scientific infrastructure
AlphaEvolve extends Gemini-based coding agents into evaluated loops for genomics, quantum simulation, infrastructure scheduling, and algorithm discovery.
AI Coding
DevTools 1.0 sets a browser QA baseline for coding agents
Chrome DevTools for agents 1.0 pulls coding agents into real browser QA and debugging loops.
Models & Research
When Flash Beats Pro, Agent Economics Take Over
Gemini 3.5 Flash is not just another fast model release. It points to the cost, latency, and routing fight behind coding agents and AI search.
AI Agents
When the search box becomes a 24-hour agent, Google tests mini apps
Google Search agents extend search from answers into persistent monitoring, task state, alerts, and generative UI mini apps.
Models & Research
Gemini API Managed Agents turn model calls into sandboxed workers
Google Gemini API Managed Agents move model calls into isolated Linux sandboxes with stateful agent execution.
Industry & Policy
Free Gemini CLI gets a deadline as terminal agents move to Antigravity
Google is moving the personal and free Gemini CLI path to Antigravity CLI. The June 18 cutoff marks a shift in the operating layer for coding agents.
AI Infrastructure
11 Seconds of Audio in Under 8 Seconds, Without a GPU
Google and Arm show how on-device generative AI is moving from model releases into CPU runtimes, quantization, memory limits, and silicon features.
Models & Research
Google AI Overviews exposes the gap behind citation cards
A May 13 arXiv study measured 55K Google searches and 98K AI Overview claims, showing where citations, ranking, and publisher economics diverge.
AI Agents
Genkit Middleware Moves Agent Control Outside the Prompt
Google Genkit Middleware shows how retries, fallback, tool approval, and filesystem boundaries are moving into the runtime layer of agent apps.