OpenAI Co-Founder Karpathy Joins Anthropic’s Pre-Training Team
▼ Summary
– AI researcher Andrej Karpathy has joined Anthropic, where he will work on pre-training and build a team focused on using Claude to accelerate pre-training research.
– Karpathy previously co-founded OpenAI, led AI at Tesla, and returned to OpenAI for a year before leaving to start the education startup Eureka Labs.
– Anthropic believes Karpathy can help it stay competitive with OpenAI and Google through AI-assisted research rather than relying solely on compute power.
– Karpathy stated he remains passionate about education and plans to resume that work in the future.
– Anthropic also hired cybersecurity veteran Chris Rohlf for its frontier red team, which stress-tests advanced AI models against severe threats.
Andrej Karpathy, the prominent AI researcher who co-founded OpenAI and once led Tesla’s autonomous driving division, has officially joined Anthropic, the company announced this week.
“I’ve joined Anthropic,” Karpathy posted on X Tuesday. “I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D.”
Karpathy began his new role this week, working on pre-training under team lead Nick Joseph. Pre-training is the massive, compute-intensive stage that gives Claude its foundational knowledge and capabilities. According to Anthropic, it is also one of the most expensive phases of building a frontier model.
An Anthropic spokesperson told TechCrunch that Karpathy will lead a new team focused on using Claude to accelerate pre-training research. This move signals that Anthropic believes AI-assisted research, not just raw compute power, is the key to staying competitive with rivals like OpenAI and Google.
Karpathy’s career has spanned some of the most influential roles in AI. He left OpenAI in 2017 to join Tesla, where he oversaw the Full Self-Driving (FSD) and Autopilot programs until 2022. He then returned to OpenAI for a year before departing again in 2024 to launch Eureka Labs, a startup dedicated to using AI assistants in education.
Since founding Eureka Labs, Karpathy has offered few updates on the venture, and it remains unclear whether he will continue with the startup. He also teaches an online course called Neural Networks: Zero to Hero, which guides students through building neural networks from scratch, and runs a YouTube channel where he posts lectures on LLMs and AI.
“I remain deeply passionate about education and plan to resume my work on it in time,” Karpathy said.
In a separate hire, Anthropic has brought on Chris Rohlf for its frontier red team, which stress-tests advanced AI models against severe threats. Rohlf, a cybersecurity veteran with more than 20 years of experience, previously worked at Yahoo’s renowned security team known as “The Paranoids” and spent six years at Meta. He also served as a fellow at Georgetown’s Center for Security and Emerging Technology, where he contributed to the CyberAI project.
“We have a real opportunity in front of us to dramatically improve cyber security with AI,” Rohlf posted on X. “I can’t think of a better company or team to join at this critical moment in time.”
(Source: TechCrunch)
