Google and Meta Lose Leading AI Researchers as New Startups Rapidly Raise Billions

Large AI companies are not just losing employees, but researchers with names and access to cutting-edge developments. More and more such specialists are leaving Google, Meta, OpenAI, and Anthropic to launch their own startups — and investors are willing to finance them almost immediately, often for hundreds of millions of dollars.

Why They Leave

The generative AI boom has dramatically raised the stakes for large labs. Companies with multibillion-dollar valuations must regularly demonstrate new models, improve benchmarks, and quickly turn research into products. Against this backdrop, strong scientists and engineers have less space for long cycles of fundamental work. Increasingly, both investors and founders themselves say that research freedom is shrinking inside corporations: priority goes to releases, commercial KPIs, and the race for leadership, not to risky but potentially breakthrough directions.

This is especially noticeable outside LLMs. When an entire organization is focused on releasing the next model, applied and more specialized research topics easily take a back seat. HV Capital partner Alexander Joël-Carbonell describes the problem directly: pressure for productivity and fast release cycles leave almost no time for real research work. For many, this becomes the turning point: if building a new lab is risky anyway, it's better to do it in your own company rather than inside a giant.

Who Has Already Attracted Funding

Examples are now appearing almost every week, and this is no longer a story about small angel checks. Rounds at new labs are more like funding for mature technology companies than seed capital for teams of a few people. What's particularly striking is that these are companies that are only a few months old: the market is buying in advance not revenue or product maturity, but the founders' reputation and the likelihood that they will assemble the next breakthrough stack.

• David Silver, former researcher at Google DeepMind, announced a $1.1 billion seed round for Ineffable Intelligence — a company just a few months old.
• Tim Rocktaschel, another DeepMind alumnus, is raising up to $1 billion for Recursive Superintelligence.
• AMI Labs, created by a Meta alumnus, received $1 billion in March 2026 just a few months after launch.
• Ricursive Intelligence, founded by former Anthropic and Google DeepMind employees Anna Goldie and Azalia Mirchoséyni, raised $335 million in two rounds just a few months after founding.

And these are no longer exceptions. Over the past year, former OpenAI, DeepMind, Anthropic, and xAI employees have collected large checks for new companies like Periodic Labs, Ricursive Intelligence, and Humans&. According to Dealroom, in 2026 alone venture investors have already invested $18.8 billion in AI startups founded since the beginning of 2025. If the pace continues, the market will exceed the $27.9 billion received by similar companies created since the beginning of 2024.

Why Investors Are Rushing

The logic of the funds is simple: in AI, value is increasingly concentrated not in the legal entity itself, but in a specific team and its ability to quickly build new models, products, or research platforms. If a startup is launched by someone who was behind notable breakthroughs at DeepMind, OpenAI, or Meta, for the market this is almost a ready-made quality stamp. Investors fear missing the next big project, so they enter at a very early stage and are willing to pay for reputation, speed, and access to rare talent.

There's also a business factor. A new independent company can sell technology to large clients without appearing as an extension of one of the tech giants. For some markets this is critical: especially where sensitive data, chips, or internal intellectual property are involved. Ricursive Intelligence cofounder Anna Goldie put it bluntly:

"To get chip manufacturers to trust us with their valuable intellectual property, we need to be like

Switzerland."

The startup's neutrality in this case becomes a commercial advantage.

What This Means

The AI market is entering a phase where the main battle is not just about models and computing, but about people capable of building a new research school around themselves. For Google, Meta, and other giants, this is a signal: retaining talent with salaries and infrastructure access alone is no longer enough. For the industry as a whole, it's a sign that the next wave of strong AI companies may be born not inside corporations, but on their way out.