Yann LeCun

New Capabilities

Left Meta; raised $1.03 billion for AI startup AMI Labs

For three years, Meta staked its artificial intelligence strategy on giving models away. The company's Llama series became the most widely used open-weight model family in the industry, downloaded hundreds of millions of times. On April 8, 2026, Meta released Muse Spark — the first model built by its new Superintelligence Labs division — and kept it closed. The shift is not subtle: the company that once argued open-source AI would defeat proprietary rivals the way Linux defeated Unix is now competing on their terms.

Updated Apr 8

New Capabilities

Departed Meta; raised $1.03 billion for new AI startup

Meta Platforms has committed more money to artificial intelligence than any other company in history—up to $135 billion in 2026 alone, and $600 billion in American data center infrastructure by 2028. But money hasn't bought capability. The company's next-generation AI model, code-named Avocado, was delayed from March to at least May 2026 after internal tests showed it trailing systems already shipping from Google, OpenAI, and Anthropic on reasoning, coding, and writing tasks.

Updated Mar 13

New Capabilities

Leaving Meta to launch AMI Labs pursuing alternative AI architectures

OpenAI's GPT-5 dropped on August 7, 2025, completing AI's transformation from chatbots that string words together to systems that actually think through problems step-by-step. Google DeepMind's reasoning models won gold at the International Math Olympiad, solving problems only five human contestants cracked. Anthropic's Claude, Meta's Llama, and every major AI lab sprinted to build models that pause, plan, and reason rather than just predict the next word.

Updated Jan 8