DeepMind's cash advantage
New reporting says Demis Hassabis and DeepMind are operating with Google’s 'war chest'—giving the lab breathing room for long‑horizon projects—while OpenAI has reportedly burned through about $14 billion and faces stronger monetization pressure. That funding gap is now framed as a recruitment and research differentiator: stability at DeepMind versus high‑velocity, commercially driven work at OpenAI. (axios.com) (benzinga.com)
Sebastian Mallaby’s new biography The Infinity Machine was published March 31, 2026 and is based on hundreds of hours of interviews with Demis Hassabis and his inner circle, framing DeepMind’s Google relationship as a deliberate platform for long‑horizon science. (penguinrandomhouse.com) DeepMind’s public job listings for Research Engineer and specialist roles explicitly require mathematical modeling, experimentation with scaling laws, and production‑grade software engineering—language that signals hires must combine ML theory, advanced math, and coding at scale. (deepmind.google) Alphabet told investors in its Q4 2025 earnings that it expects to “significantly increase spending on AI” in 2026, providing corporate capital and TPU/infra leverage that underwrites DeepMind’s multi‑year projects. (cnbc.com) Multiple outlets have reported OpenAI projecting ~\$14 billion in losses for 2026 while simultaneously shifting toward enterprise monetization, and news coverage indicates plans to nearly double headcount to roughly 8,000 employees by the end of 2026. (ibtimes.com) Even with Alphabet backing, DeepMind has lost personnel to rivals: CNBC reported Microsoft hired around two dozen DeepMind employees in mid‑2025, showing that compensation, team scope and startup‑style roles remain powerful recruitment levers. (cnbc.com) DeepMind’s publications feed and role descriptions show continued emphasis on peer‑reviewed research plus research‑engineering hybrids—favoring PhD‑level candidates who can both prove novel theory and implement large systems. (deepmind.google) OpenAI’s public research‑scientist listings require PhD‑level research ability but repeatedly stress collaboration with product and engineering teams, and industry reporting highlights growing demand for “forward‑deployed” technical ambassadorships that convert research into deployable, revenue‑driving systems. (openai.com)
