Claude Code's auto mode becomes default, raising safety stakes
Anthropic makes auto mode the default in Claude Code from August 14, claiming its classifier is safer than human approval.
Anthropic is putting auto mode in the driver's seat for Claude Code, making it the default behavior for new sessions starting August 14. The move follows months of internal testing that the company says shows its classifier is "as safe or safer than an average user clicking through prompts." The change applies to Pro, Max, and Team plans, while remaining opt-in on enterprise and cloud platforms for now.
Sources
- The Register Original source
Continue Reading
AI & MLNewHow MCP Splits Can Leak Secrets
Malicious MCP servers can split instructions to make AI coding agents exfiltrate secrets, ASSET reports.
AI & MLNewOpenAI's Daybreak Expansion Arms Defenders Against AI Threats
OpenAI expands Daybreak with new GPT-5.6-Cyber model to help defenders counter AI-driven attacks.
AI & MLAI Research Gets a Human Amplifier
PortSwigger's HTTP Terminator, guided by a human, finds novel vulnerability classes and hundreds of live targets.