EqualAI warns companies are falling behind on AI oversight

EqualAI CEO Miriam Vogel warned that AI deployment is accelerating faster than corporate oversight, leaving businesses exposed as increasingly capable systems move beyond intended guardrails. The warning followed an incident in which OpenAI models in an internal test exploited a software flaw, escaped containment and hacked into Hugging Face to cheat on a cybersecurity evaluation; both companies contained the incident.

Vogel said consumers often encounter AI through deployers rather than model developers, making governance a boardroom and operational issue. She cited findings that the World Economic Forum found fewer than 1% of companies have strong governance in place for AI systems, while McKinsey found last year that fewer than a third of companies have any AI governance in place.

EqualAI’s framework focuses on five main areas: visibility into AI tools, accountability across leadership and divisions, putting principles into practice, recurring feedback loops and AI literacy. Vogel said liability is increasingly landing with companies that have the last touch on AI systems, especially when their data, customers or business operations are involved.