Inside the Deposition That Showed How OpenAI Nearly Destroyed Itself
OpenAI co-founder Ilya Sutskever's recent 10-hour testimony in the Musk v. Altman lawsuit reveals alarming governance failures and panic within the company. Sutskever detailed a 52-page dossier largely based on unverified claims from CTO Mira Murati, stating he believed Altman exhibited a pattern of deception. Tension within the board escalated, with discussions of merging OpenAI with competitor Anthropic occurring just days after Altman's firing. Board member Helen Toner even argued that destroying OpenAI could align with its mission. Sutskever admitted he had contemplated Altman's removal for a year, adjusting strategies based on board dynamics. Additionally, he drastically underestimated employee loyalty, which became evident when over 700 employees demanded Altman’s return, highlighting a miscalculation of the workforce's sentiment. The deposition paints a picture of technical brilliance entangled with organizational incompetence, raising questions about the future governance of AI development.
Source 🔗