10 results on this page
Google DeepMind researchers found that training AI agents through debate can reduce reward hacking when an LLM judge evaluates their work. On mathematics tasks, debate achieved higher sustained ground-truth accuracy than directly optimizing for LLM judge rewards.
Google has opened Gemini features in Chrome to all Android users in the United States. The rollout follows an initial announcement in May and a limited release to selected U.S. Android devices beginning in late June, and now includes access to Auto Browse.
Google now lets users describe what they want to see in their Discover feed using natural language. The feature offers a more direct alternative to relying solely on algorithmic recommendations.
Google's AI Professional Certificate covers AI fundamentals, data analysis, and app building. The article highlights a way for learners to earn the credential without paying.
Australia’s parliament passed the News Bargaining Incentive, requiring major platforms to face the law even if they choose not to carry news. Meta, Google, TikTok, and LinkedIn are among the affected companies.
Google is offering eligible college students a free year of its paid AI Pro plan for the new school year. The company is also introducing study tools across Google Search and Gemini.
Google is working with the UK government on Operation Blue Skies to address aviation contrails with AI. The initiative will subtly reroute flights to test whether contrail formation can be avoided.
The Pixel Watch 5 refines Google’s smartwatch formula with smarter gym tracking, new health alerts, and offline Gemini support. The review presents it as a meaningful but incremental upgrade at a premium price.
Google’s Pixel 11 Pro XL offers incremental hardware upgrades, but its deeper Gemini integration is the main improvement for productivity users. The new Rambler feature converts natural, imperfect speech into cleaner written text, while Tensor G6 improves responsiveness and on-device AI performance.
Google’s Magic Capture automatically extracts photos from video footage, helping users save spontaneous moments without manually scrubbing through clips. An early test with an uncooperative cat demonstrates the feature’s potential for computational photography and everyday image capture.