Inside OpenAI’s Latest Safety Lead Exit: Is 'Iterative Safety’ A Broken Model?

The incident has become an important example in the wider debate because it raises questions about what happens when AI systems are given the ability to perform tasks independently.