Illustration by Tag Hartman-Simkins/Futurism. Source: Dimitrios Kambouris/Getty Images for WSJ. Magazine Innovators Awards
This summer, Meta CEO Mark Zuckerberg went above and beyond to revamp his company’s lagging AI division, spending billions of dollars to attract top talent across the industry. But it seems he’s already at odds with the man he hired to lead Meta’s new AI efforts, Alexander Wang.
according to new reporting From financial Times, The situation is tense between the two. Sources familiar with the matter say Wang has told colleagues that he finds Zuckerberg’s micromanagement suffocating. Meanwhile, some staffers wonder whether Wang, 28, is out of his depth, lacking both the expertise and experience to lead such a massive effort.
It’s another sign of Zuckerberg’s direction for the company, which is coming under the microscope after his last big bet — on creating a virtual reality “metaverse” — failed spectacularly. The market’s skepticism was on full display in October, when Meta announced plans to spend billions of dollars on AI this year and next, causing its stock to drop 11 percent and shed more than $200 billion in market capitalization.
His decision to hire Wang was controversial from the beginning. Wang is the founder and former CEO of AI data annotation startup Scale AI, which provides an essential service for training AI models, but does not actually build them. In June, Zuckerberg invested more than $14 billion in Scale AI, and brought Wang on board to lead Meta’s newly dubbed Superintelligence Labs. Wang is also leading a secret “TBD” (To Be Determined) laboratory that operates in his own building.
The cost of the hire wasn’t just monetary: Meta’s then-chief AI scientist, Yan LeCun, did not accept being forced to report to Wang to begin with, and suddenly went out in novemberLeCun is considered the “godfather” of the field for his pioneering work on neural networks, and he felt insulted to see his research hollowed out by his firing from a rather product-focused AI lab, while Zuckerberg offered Wang an astronomical nine-figure contract to bring talent to his superintelligence labs,
The new lab will double down on its use of large language models, the same architecture that powers AI chatbots like ChatGPT and Gemini, in its efforts to create a “superintelligence” that equals or even surpasses human capabilities. LeCun saw LLM as a dead end and instead supported creating new types of AI models.
Some employees questioned Wang’s credentials to lead such a significant effort, and it’s a fair issue to raise since Wang’s company hasn’t built any AI models at all. Instead, its focus was on the data used to train them, an entirely different kettle of fish. Now, they are expected to develop not just AI models, but models that will rival human intelligence.
This is not the first sign of friction with Wang at the company. new York Times informed Earlier this month, Wang was at loggerheads with Zuckerberg’s longtime lieutenant Chris Cox, who wanted to focus on using Facebook and Instagram to train Meta’s new models. Wang disagreed and argued to instead focus on capturing Google and OpenAI’s models.
Zuckerberg’s determination to build these products as fast as possible will only increase the tension. “If you build too slowly,” he said in an interview on the Access Podcast foot“Then you’re out there with what I think is going to be the most important technology that enables the most new products and innovation and value creation in history.”
His other big new hire, former GitHub CEO Nat Friedman, is in a similar position to Wang, and some on his team were disappointed after the hasty release of Vibes, a feed of AI-generated videos similar to OpenAI’s Sora 2 app.
The pressure is not going to ease any time soon. The secretive TBD Lab aims to release an entirely new AI model built from scratch in the first quarter of next year, people familiar with the matter said. foot,
More on Meta: Metra’s $27 billion datacenter is wreaking havoc on Louisiana city
