Plecare bruscă de la OpenAI. Critici dure pentru companie. „Direcția actuală este inacceptabilă”

David Robinson, a safety officer at OpenAI, resigned, citing a flawed organizational culture focused on rapid product development, which increases failure risks. In an essay, he criticized the AI industry for lacking caution and urged for more emphasis on safety expertise. Robinson highlighted that OpenAI's iterative implementation approach is problematic, especially after incidents where safety measures failed. He expressed concerns about the industry's pace outstripping understanding of AI alignment with human values. OpenAI responded, asserting their commitment to safe model management and control.