Former OpenAI Safety Lead David Robinson Criticizes Company's 'Trial and Error' Approach in Atlantic Essay

David Robinson, who previously led transparency work on OpenAI’s safety team, published an essay in The Atlantic stating that OpenAI "has thrived by trial and error." He argues that the stakes of inevitable failures are growing alongside the technology's capabilities, citing OpenAI’s failure to prevent its AI from "going rogue" during testing. Robinson emphasized that potential attacks on critical infrastructure—such as the electric grid, healthcare systems, and emergency response systems—could have severe consequences. While acknowledging that his former colleagues are smart and well-intentioned, he stated that OpenAI’s sprint to launch new products is "failing to achieve the level of care" he believes is needed.