How much would you have to get paid to spend any amount of time with Mark Zuckerberg? For some of the folks who have joined Meta’s AI team, millions of dollars simply is not enough. According to a new report from the Financial Times, the Meta AI lab is such a mess (and presumably, Zuck is such a bad hang) that recent hires have threatened to quit within days of joining, and promotions have been handed out just to keep people from leaving.
FT points to Shengjia Zhao, Meta’s “chief AI scientist,” as a prime example of this dynamic. How’d Zhao end up with that title? Well, the former OpenAI employee who was involved in the development of ChatGPT spent a few days within Meta’s operation and presumably saw what a shitshow it was. He immediately started making arrangements to leave, going so far as to sign paperwork to return to OpenAI, according to the Financial Times, before Meta agreed to hand over the high-ranking title. And they say Silicon Valley isn’t really a meritocracy!
Zhao is apparently far from the only one of Meta’s very high-profile hires who has shuddered at the sight of the Superintelligence Lab. Earlier this week, Wired reported that multiple recent additions to the company’s AI effort have already abandoned ship. That included former OpenAI employees Ethan Knight and Avi Verma, and Google DeepMind alum Rishabh Agarwal, all of whom spent only a handful of months at Meta’s operation. Veterans of Meta are getting out, too, with more than two dozen long-time staffers departing in recent weeks, according to FT.
Frankly, it’s hard to blame them if you’ve been following the headlines out of Meta’s AI efforts these last few months. The company has undergone reorganization after reorganization. Currently, it’s on its fourth major overhaul in just over six months. The company has thrown money at the problem, offering massive pay packages to the biggest names in the field as Meta tries to play catch-up with the likes of OpenAI and Anthropic, as well as more established players like Google. Zuckerberg was apparently blinded by his VR goggles when he was all-in on the metaverse to see the AI revolution coming, but all his efforts to make up for lost time seem to just put the company further behind the eight ball.
The situation has reportedly not been improved by the presence of Alexandr Wang, the former CEO of Scale AI, who was absorbed into the role of Chief AI Officer of Meta when Zuck poured $15 billion into Scale to get his hands on their talent and infrastructure. Wang, though, has apparently lived up to his name and is a real dick, clashing with Zuck and rubbing employees the wrong way, according to FT.
The benefit of having the infinite money glitch, as Meta does, is that most problems can be fixed by buying up as much talent as possible and figuring the rest out later. Unfortunately for Zuck, while he can afford any AI expert in the field, he’ll never be able to buy a new personality.
