San Francisco (GNP): A former OpenAI safety employee who recently resigned has criticised the company’s approach to artificial intelligence (AI) safety. David Robinson argues that a fast-paced culture built around rapid development raises the risk of failures.
He made the case in an essay titled “I Quit OpenAI Because Its Culture Is Broken,” published on Saturday. Robinson wrote that AI companies, OpenAI among them, are not being nearly careful enough. In his view, they should put greater emphasis on safety expertise and research before building more capable systems.
His central demand is a change in standards. Robinson argued that advanced AI systems need safeguards closer to those used in industries such as nuclear power and aviation.
He pointed to OpenAI’s own method as the source of the problem. According to Robinson, the company relies heavily on what it calls “iterative deployment,” which means releasing systems and strengthening safeguards when problems emerge.
His warning carries weight because of his role inside the company. Robinson said he spent three and a half years at OpenAI. He helped draft its preparedness framework and oversaw safety reports for 12 frontier-model launches.
That experience shaped his verdict on the company’s pace. He wrote that as the company sprints from one launch to the next, it is failing to achieve the level of care he believes is needed.
Also Read: AG600 Amphibious Aircraft Ends Four-Month Fire Mission
OpenAI has rejected the framing. A company spokesperson said OpenAI is making sure its models do not become more capable than it can safely manage and secure. The spokesperson added that the company pauses training or holds back models when it needs to slow down.
The exchange adds to a wider debate in the AI industry over whether companies are moving too quickly to develop increasingly powerful systems. OpenAI and its rival Anthropic have both faced scrutiny after incidents in which safety controls failed or experimental systems behaved unexpectedly.
Robinson also raised a deeper concern. He warned that AI capabilities are advancing faster than researchers’ understanding of alignment, the field focused on ensuring AI systems act in line with human goals and values.
Robinson said, “The time for trial and error is over.”
