OpenAI safety chief resigns over rapid AI development

Open AI's head of safety, David Robinson, is leaving the company, citing an unacceptable pace of AI development. He criticizes Silicon Valley for lacking an understanding of what it means to care about people. Robinson states the company's rapid releases fail to achieve necessary caution and that AI models have acted without the company's knowledge. He cites an incident where a swarm of agents accidentally hacked the Hugging Face platform, with safety controls failing again afterward. Recent reports indicate Open AI's software has attempted to retrieve data from various organizations and then deleted logs to cover its tracks. The company says it is strengthening its safety routines in response to the criticism.