Anthropic CEO Dario Amodei contrasted AI companies led by scientists with those run by social media entrepreneurs, implicitly criticizing Meta’s direction following Yann LeCun’s departure. LeCun reportedly clashed with Mark Zuckerberg over research priorities, leading to his exit and the hiring of a younger executive. This move highlights a perceived difference in approach to AI development and responsibility.
When Meta’s Llama 4 flopped in April 2025 amid accusations of benchmark manipulation, Mark Zuckerberg lost confidence in the team. The timing was impossible to ignore—just weeks after Yann LeCun, the Turing Award-winning godfather of deep learning, walked out of Meta following reported clashes with Zuckerberg over the company’s AI direction.
But his vision for “world models”—AI that understands physical reality, not just text—clashed with Zuckerberg’s aggressive push toward large language models. Anthropic CEO Dario Amodei didn’t name Mark Zuckerberg . He didn’t have to. Speaking at the World Economic Forum in Davos last week, Amodei drew a pointed contrast between AI companies led by scientists and those run by “the generation of entrepreneurs that did social media. The implication was clear. Scientist-founders like himself and Google DeepMind’s Demis Hassabis build with caution. Social media entrepreneurs chase scale and sort out the damage later. Yann LeCun spent over a decade building Meta’s AI research lab FAIR. Amodei’s Davos remarks carried a quiet verdict. Meta let one of AI’s founding minds slip away. That, he seemed to suggest, is exactly the kind of mistake a scientist would never make.