AI still can’t run reliable end-to-end cyberattacks without humans, report finds
The International AI Safety Report 2026 says AI can automate parts of the cyberattack chain but still fails at long, multi-stage execution.
The International AI Safety Report 2026 says AI can automate parts of the cyberattack chain but still fails at long, multi-stage execution.
The VP of product on MCP servers, legacy data modernization and what her English degree taught her about building for developers