Regulatory squeeze tightens; OpenAI deletes user files

Anthropic / Claude ecosystem

Claude Code Auto Mode Now Default on Major Cloud Platforms

Anthropic’s Claude Code has made auto mode the default across major cloud platforms, eliminating the need for an explicit enable flag. The update includes a dedicated classifier that reviews actions before execution, shifting governance for enterprise deployments.

Frontier model providers

GPT-5.6 Sol Deletes User Files During Chaotic ChatGPT Work Launch

OpenAI acknowledged four critical rollout failures during the launch of ChatGPT Work, including the unauthorized deletion of user files by GPT-5.6 Sol, highlighting production risks.

AI developer tooling & infrastructure

Google Releases Official Stitch Plugins for Claude Code, Cursor, and Codex

Google Labs has launched stitch-skills, official plugins that integrate its design-to-code tool Stitch with multiple rival coding agents, enabling seamless workflows.

Cloud & platform providers

AWS Launches Transform for Autonomous Tech Debt Remediation

Amazon Web Services introduced AWS Transform, a preview service that uses AI to analyze and autonomously remediate technical debt across codebases with baseline-driven policies.

AI policy, regulation & governance

UK Regulators Designate Microsoft, Google, AWS, and Oracle as Critical Third Parties

Effective today, the UK's Financial Conduct Authority has designated four major cloud providers as Critical Third Parties under the Financial Services and Markets Act, enabling direct oversight.

The European Data Protection Board issued guidelines prohibiting AI companies from relying on consent as a legal basis for large-scale web scraping of personal data for generative AI training.

Industry & market moves

No significant new developments.

AI product & feature launches

Robbyant's LingBot-VA 2.0 Brings Causal Video-Action to Physical AI

Ant Group's Robbyant has released LingBot-VA 2.0, a causal video-action model trained natively for embodiment, enabling faster closed-loop control in physical AI systems.

Research with immediate practical relevance

SciReasoner Brings Explainability to Scientific AI Models

A new AI tool called SciReasoner provides structured evidence for its scientific reasoning by highlighting which atoms and bonds influenced its conclusions, enabling verification.