Why AI agents go rogue — and what it means for CFOs

Incidents disclosed by OpenAI and Anthropic expose how AI models can behave in unintended ways, creating new challenges for finance chiefs.

Published: 2026-10-20