Anthropic has released findings from a pilot program that allowed three outside research groups to study aggregated data from roughly 250,000 Claude and Claude Code conversations. One study found that users delegate consequential tasks to the AI more often than earlier research had suggested, especially when seeking professional legal or financial guidance. Kunal Handa and …
OpenAI is testing a “Persistent mode” for its AI coding agent in Codex that could allow it to continue tasks until a user tells it to stop. Maxwell Zeff reports for WIRED that references to the feature have appeared in public changes to Codex’s command line code base. The setting is not publicly available, and …
Google is adding a feature called Expert Intelligence to Gemini Notebook that lets users import eligible books they own through Google Play Books. The feature turns purchased titles into sources that users can question, summarise and reuse in AI-generated formats. Emma Roth reports for The Verge that users can also create plans, infographics and AI-generated …
Anthropic is rolling out a built in browser for Claude Cowork that lets the AI open websites, read pages, click controls and enter text without leaving the desktop app. The feature is intended for tasks such as completing web forms or retrieving figures from dashboards that do not offer an API. The browser opens in …
Google has introduced Gemini Omni 1.1 Flash, an update to its generative video model aimed at giving developers and creative teams more control over how clips are made, extended and refined. The company is adding scene continuation, keyframe-based video generation, low-cost draft rendering, video references and output options up to 4K. The update addresses a …
Nvidia is reportedly acquiring AI platform Hugging Face for $12.9 billion, a deal that could give the chipmaker a far stronger position in the market for openly available AI models and related cloud services. Matthias Bastian reports for The Decoder that the information comes from The Information. Neither company is cited as confirming the transaction. …
Google is expanding Gemini Live with voice-driven task automation, daily summaries and Gmail controls. In an official Google Blog post, Neel Joshi announces that the update connects Gemini Live with Spark, Google’s system for handling longer, multi-step tasks. The new Spark integration lets users describe a goal by voice and have Gemini carry out related …
Alibaba has introduced Qwen3.8-Flash-Next, a multimodal AI model designed to deliver strong coding and workplace performance while using far less computing power than larger rivals. Matthias Bastian reports for The Decoder that the release also previews architectural ideas Alibaba plans to bring to its future Qwen4 model family. Qwen3.8-Flash-Next contains 125 billion parameters overall, but …
Google has introduced Gemini 3.5 Transcribe, a speech-to-text model for live conversations, recorded meetings and voice-driven software. The company says the model produces formatted text directly from raw audio and is built to handle noise, specialist vocabulary, accents and spoken corrections. The release comes in two versions for developers. The Live API model, called gemini-3.5-transcribe-live, …
OpenAI CEO Sam Altman says the company expects to have an internal system he would classify as artificial general intelligence, or AGI, by the end of the year. The claim comes as OpenAI slows parts of its research after an unreleased agent escaped a testing environment and accessed systems at AI platform Hugging Face. In …