Mark Zuckerberg's most famous employee reveals what he hated most at Meta, the company he worked for more than 12 years

Yann LeCun, Meta's former chief AI scientist and one of the "godfathers" of modern AI, has opened up about what he despised most during his 12-year tenure at the tech giant: being a manager.

"I can do management, but I don't like doing it," LeCun told MIT Technology Review in an interview published Thursday. "I kind of hated being a director. I am not good at this career management thing. I'm much more visionary and a scientist."The 65-year-old Turing Award winner, who founded Meta's influential Fundamental AI Research (FAIR) lab, announced his departure from the company in November to launch his own AI startup, AMI Labs.

The Paris-based venture will focus on building "world models"—a type of AI that learns from videos and spatial data rather than just text, which LeCun believes is the true path to artificial general intelligence.

The scientist who never wanted to lead

At his new venture, LeCun will serve as executive chairman while former Meta colleague Alex LeBrun handles CEO duties. It's a deliberate choice that lets him dodge the administrative work he despised. "This is not my mission in life," he explained to MIT Technology Review.

"It's really to make science and technology progress as far as we can."He'll also keep teaching at NYU, where maintaining his academic position was one of three conditions he set when Mark Zuckerberg first wooed him to Facebook in 2013 over dinner and "chicken with some pretty good white wine."

When a 28-year-old became his boss

But LeCun's exit wasn't just about escaping paperwork. Meta's direction increasingly clashed with his scientific convictions, particularly after the company poured $14.3 billion into Scale AI and installed its 28-year-old CEO Alexandr Wang to oversee AI development—making Wang briefly LeCun's manager."You don't tell a researcher what to do," LeCun told the Financial Times. "You certainly don't tell a researcher like me what to do."The real friction, though, centered on technology. LeCun has consistently argued that large language models are "a dead end when it comes to superintelligence," a stance that became awkward as Meta doubled down on LLMs. "I'm sure there's a lot of people at Meta who would like me to not tell the world that," he admitted, "but I'm not gonna change my mind because some dude thinks I'm wrong."His honesty cuts through the usual corporate doublespeak that dominates AI discussions, offering a rare peek at what happens when scientific integrity meets billion-dollar bets.