The AI industry’s talent war reached a new inflection point on Tuesday. Andrej Karpathy, one of the most prominent researchers in artificial intelligence and a founding member of OpenAI, announced that he has joined Anthropic. The move lands at a moment when American AI labs are spending unprecedented sums to attract the technical leaders shaping the next generation of large language models.
Karpathy made the announcement on X on May 19, writing: “Personal update: I’ve joined Anthropic. 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. I remain deeply passionate about education and plan to resume my work on it in time.”
The news was confirmed across major outlets including Axios, Bloomberg, CNBC, and TechCrunch within hours of Karpathy’s post.
Where Karpathy Lands
According to Anthropic, Karpathy began his new role this week on the company’s pre-training team, reporting to team lead Nick Joseph. Pre-training is the foundational stage in which large language models acquire their core knowledge and capabilities through massive training runs on vast datasets. It is also one of the most resource-intensive and strategically important phases of building a frontier model, with improvements at this stage producing outsized effects on a model’s performance, efficiency, and capability scaling.
An Anthropic spokesperson told TechCrunch that Karpathy will help launch a new team focused on using Claude itself to accelerate pre-training research. The premise — letting an AI model help train its successors — represents one of the more closely watched frontiers in the field, and one Anthropic has signaled it intends to push aggressively.
In a separate report cited by Axios, Anthropic framed the hire as a step toward making AI-assisted research, rather than pure compute scale, the core of how the company plans to compete with OpenAI and Google.
A Long Path Through the AI Industry
Karpathy’s career has tracked the major inflection points in modern AI. After helping start OpenAI, he left in 2017 to join Tesla, where he led the company’s Autopilot and Full Self-Driving programs. He departed Tesla in 2022, returned to OpenAI for roughly a year, and then left again in 2024 to start Eureka Labs, a startup applying AI assistants to education.
He holds a PhD in computer science from Stanford and has built a substantial following through his Neural Networks: Zero to Hero online course and his YouTube channel, both of which have helped train a generation of developers in the technical foundations of modern LLMs. It remains unclear whether Karpathy will continue his work with Eureka Labs alongside his role at Anthropic; his X post indicated he plans to return to education work “in time.”
His name surfaced repeatedly during the recently concluded Musk v. Altman trial, which centered on the early history of OpenAI and concluded on Monday, per CNBC.
Part of a Larger Talent Push
Karpathy’s arrival is the most visible in a series of senior moves at Anthropic. Earlier in May, Ross Nordeen, a founding member of Elon Musk’s xAI and a former Tesla employee, joined Anthropic. The announcement came the same day Anthropic struck a deal with SpaceX to rent compute capacity at xAI’s Colossus 1 data center in Memphis, Tennessee, per CNBC.
The two developments — pulling senior AI talent from xAI and Tesla while securing compute capacity from a Musk-affiliated facility — underscore the complicated commercial geography of the AI industry, where rival labs share contractors, infrastructure, and increasingly, people.
Anthropic is in fresh financing discussions and is reportedly poised to surpass OpenAI’s private market valuation, per Bloomberg, with elite hires functioning as both research investments and recruiting signals. Founded by former OpenAI researchers Dario and Daniela Amodei, Anthropic has positioned itself as an AI safety-focused alternative to OpenAI, and the past year has shown how aggressively the company is willing to compete at the frontier.
Implications for the Broader Tech Labor Market
For the wider U.S. tech labor market, the Karpathy hire is the kind of marquee move that influences thousands of downstream decisions. Senior researchers signal where the frontier is forming, and their relocations shape where junior engineers, applied scientists, and product talent want to go next.
Compensation for senior AI researchers has continued to climb. TechCrunch reported earlier this year that several startups are paying tech-savvy graduates over $300,000 to fill roles that agent-driven platforms have not fully automated. At the senior level, total compensation packages for researchers of Karpathy’s profile reportedly run into the tens of millions, though Anthropic has not disclosed terms.
The hire also reinforces the perception that the AI industry’s leading scientists move in tighter circles than the broader tech sector. The same names — Anthropic, OpenAI, Google DeepMind, xAI, Meta’s superintelligence team — keep appearing in transition announcements, and the bench supporting frontier model development remains, by most accounts, surprisingly small.
For Anthropic, Karpathy joins at a moment when the company’s research roadmap, financing, and compute footprint are all expanding at once. For the industry, the announcement is another reminder that the AI talent race is far from settling.
