Ex-OpenAI Safety Staffer Criticizes Company Culture: The Era of Trial and Error Is Over

A former OpenAI safety staffer who recently resigned, David Robinson, criticized the company's approach to AI safety in an article published in The Atlantic. He said the company's high-pressure culture of chasing rapid development raises the risk of system failures, and AI companies should value safety talent and research more before building stronger systems.

Robinson said OpenAI relies on an "iterative deployment" model, releasing systems first and strengthening safety later. He worked at OpenAI for three and a half years, helped draft its risk response framework, and oversaw safety reports for 12 frontier model releases. He warned that AI capabilities are advancing faster than researchers' understanding of alignment.

An OpenAI spokesperson said the company keeps model capabilities within safely manageable limits, pausing training or delaying releases when needed. The remarks have further fueled industry debate over whether development is moving too fast.