Anthropic's Rumored Physical Intelligence Bid Raises Stakes in AI-Robotics Consolidation Race

Over the weekend, word spread rapidly across social media that Anthropic was acquiring Physical Intelligence, the well-funded robotics startup co-founded by prominent Silicon Valley investor-operator Lachy Groom. While Physical Intelligence's leadership issued a soft denial, it turns out the rumor had some grounding in reality: the two companies actually held acquisition discussions earlier this year, according to reporting from The Information.

Why This Rumor Stuck#

Physical Intelligence isn't just another robotics venture. The startup has raised over $1 billion, was recently valued at $11 billion, and developed the π0.5 model—a robot brain now widely adopted across robotics research. Groom's rising profile in the startup ecosystem, combined with the company's connections to top-tier investors like Khosla Ventures and Thrive Capital, gave the acquisition rumors credibility.

Physical Intelligence CEO Karol Hausman's response to employees via Slack—reportedly a gif of a head shake from The Office—was notably understated, lacking the forceful pushback a company might typically deploy against acquisition speculation. Meanwhile, Groom declined to comment when contacted.

The Bigger AI Consolidation Picture#

Anthropic and OpenAI are on major acquisition sprees, betting that acquiring developer tools, AI service firms, and robotics talent will accelerate their paths to greater market reach and enterprise revenue. Anthropic has completed four acquisitions this year; OpenAI has been more aggressive, with at least 17 deals since 2023. Both companies have now filed confidentially for IPOs—Anthropic in June, OpenAI a week later—suggesting both see acquisitions as critical to their valuations.

Why Robotics Matters Now#

The prevailing logic in AI circles is that physical-world understanding may become essential for building superintelligent systems. Text data alone has limits. OpenAI's own trajectory illustrates this thinking: the company once built a robotic hand that could solve a Rubik's Cube, then shut down its robotics team in 2021 after co-founder Wojciech Zaremba concluded the work lacked pieces critical to true superintelligence. OpenAI quietly restarted its robotics efforts in 2024 with a new humanoid lab in San Francisco, then CEO Sam Altman made it public in May, framing near-term infrastructure robots as stepping stones toward a personal robot for every consumer.