Insights
Field notes on operationalizing AI, from readiness to governance to what happens when things go wrong.
What We Learn When AI Goes Rogue
This summer something the AI governance community had theorised about for years actually became reality: AI agents broke containment. Whether you’re just starting to get serious about AI or your teams are already integrating it into their daily operations, there are major lessons that can be applied to your organization.
OpenAI's Shutdown Capability Is the Right Move But It Is Not Enough
True AI safety cannot rely solely on intervention after-the-fact, or self-reported information; it requires clear, independent mechanisms to ensure human oversight and embedded accountability. Read more in our researcher’s opinion piece.
The Illusion of Plug-and-Play: The Pizza Hut Lawsuit
When you think about potential enterprise AI failures, you might think of model hallucinations, biased data, coding issues, or security vulnerabilities. But what about the crises caused by software and workflows that technically were executed according to plan? What those incidents often lack isn’t quality assurance testing but adequate scoping.
The Boardroom Gap: Why AI Strategy is Failing at the Foundational Level
With the release of GenAI and agentic AI, the technology has exited the lab and entered the boardroom at a scale we’ve never seen. It has moved from "specific use" to "general use," released into the wild of everyday work and life, where it is significantly harder to control.