A former OpenAI safety employee has criticised the company’s approach to AI development. David Robinson argued that a fast-paced culture focused on rapid releases could increase the risk of serious failures.

Robinson made the comments in an article titled I Quit OpenAI Because Its Culture Is Broken, published by The Atlantic on Saturday.

Call for Stronger Safety Measures

Robinson said AI companies, including OpenAI, are not being careful enough. He urged them to give greater weight to safety expertise and research before developing more capable systems.

“The time for trial and error is over,” Robinson wrote. He argued that advanced AI systems need safeguards comparable to those used in high-risk industries such as nuclear power and aviation.

Cristiano Ronaldo Jorge Jesus Leaves Door Open for Portugal Return

According to Robinson, OpenAI relies heavily on “iterative deployment”. The approach involves releasing systems and strengthening safeguards when developers identify problems.

Debate Over AI Development

His comments add to a wider debate within the AI industry. Researchers and safety experts have questioned whether companies are moving too quickly as they develop increasingly powerful systems.

OpenAI and rival Anthropic have also faced scrutiny following incidents involving safety controls and unexpected behaviour from experimental systems.

Robinson said he spent three and a half years at OpenAI. During that period, he helped draft the company’s preparedness framework and oversaw safety reports for 12 frontier-model launches.

“As the company sprints from one launch to the next,” Robinson wrote, “it is failing to achieve the level of care that I believe is needed.”

OpenAI Defends Its Approach

OpenAI rejected the criticism and said it takes safety risks seriously.

“We’re making sure our models don’t become more capable than we can safely manage and secure,” an OpenAI spokesperson said. The company added that it pauses training or holds back models when it needs to slow down.

Robinson also warned that AI capabilities are advancing faster than researchers’ understanding of alignment. The field focuses on ensuring AI systems follow human goals and values.

The debate highlights a central challenge for the AI industry: how to advance increasingly capable systems while keeping safety research and safeguards in step with their development.