A curious migration is underway in the tech world. Entrepreneurs and executives who have already amassed fortunes and built household-name companies are quietly returning to the coalface — not as investors or board members, but as individual contributors in artificial intelligence startups.
The latest example is Tom Blomfield, co-founder of Monzo and GoCardless, who announced on Monday that he is taking a leave of absence from his role as a Y Combinator partner to join Anthropic's compute team. He will hold the title of 'Member of Technical Staff' — a deliberately flat designation used by both Anthropic and OpenAI to signal that even the most experienced hires are expected to code. He joins Instagram co-founder Mike Krieger, who became Anthropic's chief product officer in 2024, and Andrej Karpathy, a founding OpenAI member who joined the company's pre-training team in May.
The pattern extends beyond joining existing labs. Chamath Palihapitiya, the billionaire investor known for his SPAC deals and the 'All In' podcast, has taken his first full-time operating role in over a decade as CEO of 8090 Labs, an enterprise AI coding startup. The company announced a $135m Series A led by Salesforce Ventures a few weeks ago. Palihapitiya wrote on X that he is convinced what he is building now 'is even more important' than his previous ventures. Similarly, Eric Wu, who ran property tech firm Opendoor for a decade, recently launched NavigateAI, an AI 'copilot' for construction workers, backed by $25m in seed funding. Wu told this publication he knew he would regret it if he looked back in ten years and had not done something related to AI.
For UK businesses and consumers, this trend signals that the AI sector remains in its early, high-stakes phase. When serial founders and former CTOs of billion-dollar companies are willing to accept junior-level titles to work on frontier models, it suggests the potential disruption — and reward — is seen as extraordinary. The UK's Information Commissioner's Office (ICO) has been navigating a balancing act between fostering AI innovation and enforcing data protection rules under the UK GDPR, while the EU AI Act imposes a tiered regulatory framework that will affect any UK company selling into Europe. The movement of top talent into these labs may accelerate the pace of product releases, putting pressure on British firms to adopt AI tools or risk falling behind.
Experts warn of both opportunity and risk. The concentration of elite talent in a handful of labs — particularly Anthropic and OpenAI — raises questions about market dominance and the ethical guardrails around powerful AI systems. For UK startups competing for engineers, the gravitational pull of these US-based frontier labs could worsen an already tight talent market. However, for UK consumers, the influx of experienced builders may lead to more practical, safer AI products reaching the market sooner. As Blomfield put it, the next few years at the frontier of large language models 'will be especially formative' — a sentiment that is clearly shared by a growing number of people who have already won the game.