The AI 2027 Tracker, a project tracking predictions from the AI 2027 scenario, reports that several key milestones have been confirmed or are emerging, including the completion of a frontier model tra…
Google integrated 'Computer Use' into Gemini 3.5 Flash, enabling the AI model to operate computers, browsers, and mobile devices autonomously. The model achieved a score of 78.4 on the OSWorld benchma…
Google merged computer use capabilities directly into Gemini 3.5 Flash on June 24, eliminating the need for a separate model. Developers can now enable screen automation by adding a single tool parame…
Researchers introduced OSGuard, a benchmark suite for evaluating safety in computer-use agents under benign user instructions. The suite includes an action-level benchmark for local guardrail decision…
Anthropic shipped Claude Sonnet 4.6 in February 2026, five months after Sonnet 4.5, with the same API pricing of $3 input and $15 output per million tokens. Sonnet 4.6 achieved an 80.2% SWE-bench Veri…