Top AI researcher Rishabh Agarwal quits Meta’s Superintelligence Lab despite million-dollar pay

Rishabh Agarwal, a prominent AI researcher, has left Meta’s Superintelligence Lab (MSL) just months after joining on a million-dollar salary. Sharing the news on X, he wrote, “This is my last week at @AIatMeta. It was a tough decision not to continue with the new Superintelligence TBD lab, especially given the talent and compute density.”

After spending over seven years at top tech firms such as Google Brain, DeepMind, and Meta, Agarwal said he is ready to take “a different kind of risk”.



He said that despite receiving a strong pitch from Meta CEO Mark Zuckerberg and chief AI officer Alexandr Wang, he chose to follow Zuckerberg’s own advice: “In a world that’s changing so fast, the biggest risk you can take is not taking any risk.” Agarwal also stated that he enjoyed working with his team at Meta.

After passing out of IIT Bombay with a background in computer science and engineering, Agarwal earned his PhD in AI at Mila – Quebec Artificial Intelligence Institute, in Canada.

Reflecting on his stint, Agarwal said, “In my short time at Meta, we did push the frontier on post-training for ‘thinking’ models.”

When Mark Zuckerberg launched MSL and started hiring aggressively earlier this year, it was positioned as a bold move to catapult the company forward in the AI race. Meta reportedly offered generous pay packages to poach top talent from competitors, including OpenAI, DeepMind, and Apple.

According to The Verge, total annual compensation for AI roles at Meta has ranged between $1 million and $1.4 million, including base salary, bonuses, and stock grants. However, Zuckerberg, during an interview in July, emphasised that a lot of the numbers that have been reported are “inaccurate”.

Whatever the figure may be, for some, the handsome pay package does not appear to matter. Aside from Agarwal, other researchers, including Avi Verma and Ethan Knight, both formerly with OpenAI, have also left MSL. They are now returning to their previous employer.

While the exact reasons behind Agarwal’s departure may remain unclear, one engineer who turned down Meta’s offer told The Verge that the company expected significant “personal sacrifices” in return for the high salaries. These included compromises on work-life balance and, in some cases, on individual values regarding AI development.

Such expectations appear to be discouraging not only potential recruits but also new hires.