[openai-blog] Team update
OpenAI published a team update on 30 January 2017 announcing the departure of Ilya Sutskever, co-founder and research director, and Andrej Karpathy, founding member [source]. Both researchers left to pursue independent projects.
Sutskever had served as research director since OpenAI's founding in December 2015. Karpathy joined as a founding member and contributed to the organisation's early research direction. The announcement provided no detail on succession planning or reassignment of their responsibilities.
The update stated that both departures were amicable and that the individuals would remain advisors. No interim appointments were named. OpenAI's research leadership structure at the time was not disclosed in the post.
The timing coincided with OpenAI's transition from a pure research laboratory toward product development. The organisation had recently begun exploring commercial applications of its models, though no products had yet been released to the public.
Staff departures at AI providers can affect model development timelines and research priorities. Leadership changes in research organisations have historically preceded shifts in technical direction, though no such shift was announced in this case.
The post did not address whether ongoing research projects would be affected or whether external collaborations led by either researcher would continue. OpenAI's team page was updated to reflect the changes, removing both individuals from active staff listings.
This personnel change occurred approximately 14 months after OpenAI's founding and preceded the organisation's later restructuring into a capped-profit entity by more than two years.
Why this is an AI incident
Launch-archive bulk classification (10 May 2026). Source signal originates from a real AI provider, regulator, or model-comparison probe; the harm or behavioural change described would not have occurred without the AI system being deployed in the role described. Editor reviewing the archive may amend the rationale per-wire.
Counterfactual "but-for" test per the Editor's Guide.