Andrej Karpathy, a founding member of OpenAI and a former key researcher for the company until February, announced on Wednesday (July 17) his plans to launch Eureka, an “AI native” school that integrates education and artificial intelligence (AI).
In a post on X (formerly Twitter), Karpathy highlighted the scarcity of subject matter experts who are passionate, skilled in teaching, infinitely patient, and fluent in all languages. He explained that recent advancements in generative AI have made it feasible to enhance the learning experience. In his vision, teachers would still design course materials but would be supported and scaled by an AI Teaching Assistant optimized to guide students.
Bloomberg News, which reported the announcement, noted Karpathy’s significant role in the AI sector. After helping to start OpenAI, he worked at Tesla as the senior AI director overseeing the autopilot team before returning to OpenAI and later resigning to pursue personal projects.
Karpathy’s new venture follows the launch of Safe Superintelligence (SSI) by another former OpenAI executive, co-founder, and former chief scientist Ilya Sutskever. SSI aims to create a safe and powerful AI system, reflecting an evolution from OpenAI’s original mission.
OpenAI, co-founded by Sutskever in 2015, seeks to develop artificial general intelligence (AGI) that can rival human abilities and benefit all of humanity. The organization comprises the non-profit OpenAI, Inc. and its for-profit subsidiary OpenAI Global, LLC. OpenAI has been instrumental in the current AI boom, developing technologies like DALL·E and ChatGPT, and has received significant investments from Microsoft.
Despite its success, OpenAI has faced criticism for shifting towards a more commercial focus, with some experts arguing that it has deviated from its original mission of developing safe and beneficial AGI.