Prabowo Widodo
The year has been marked by a relentless pace of AI acquisitions, a flurry of strategic moves so frequent that many barely register beyond industry circles. Giants like Anthropic and OpenAI have been on aggressive buying sprees, rapidly assimilating developer tooling firms, AI services providers, and product-testing startups. This intense M&A activity serves a dual purpose: converting advanced model capabilities into lucrative enterprise revenue streams and extending their market reach at a speed designed to outmaneuver competitors. Amidst this ceaseless churn, a weekend rumor about Anthropic’s potential acquisition of robotics startup Physical Intelligence rapidly captured attention, spreading like wildfire even after an initial denial from Physical Intelligence’s CEO.
The extraordinary speed and resonance of this particular rumor stemmed largely from the high-profile entities involved. Physical Intelligence is far from an obscure player in the nascent field of robotics. It boasts a formidable pedigree, co-founded by Lachy Groom, an investor-operator whose influence and star have been steadily ascending within Silicon Valley’s intricate ecosystem in recent years. The company’s financial backing is equally impressive, having already secured more than $1 billion in funding. Reports earlier this spring indicated that Physical Intelligence was in discussions for yet another substantial funding round, aiming to raise an additional $1 billion at a staggering $11 billion valuation. Furthermore, its "Iota 0.5 model" is widely recognized as one of the more prevalent and impactful robot brains in contemporary robotics research, underscoring its technological significance and market potential.
As subsequent investigations revealed, the initial rumor, though swiftly denied, was not entirely baseless. Confidential sources, including a detailed report from The Information, confirmed that Anthropic and Physical Intelligence had indeed engaged in acquisition discussions earlier this spring. This revelation suggests that tech blogger Robert Scoble, whose post on X ignited the speculative frenzy over the weekend, may have erred on specific details but was fundamentally correct in identifying that significant interactions had taken place between the two companies. The initial response from Physical Intelligence to the swirling rumors also warranted closer scrutiny. It was notably not a particularly forceful or unequivocal denial. According to The Information, Physical Intelligence CEO Karol Hausman communicated to employees via a Slack message that the reports were untrue, accompanying his statement with a somewhat ambiguous GIF of a character from "The Office" shaking her head in a dismissive "no." Lachy Groom, for his part, remained silent, not responding to requests for comment sent by TechCrunch on Monday night.
The broader context of this rumored acquisition illuminates the escalating arms race between leading AI developers. Anthropic has publicly acknowledged four acquisitions this year, signaling a clear intent to expand its capabilities. OpenAI, its primary rival, has been even more assertive, having acquired a minimum of 17 companies since 2023. These aggressive acquisition strategies are not isolated maneuvers; they are integral components of a larger plan. Both companies are now meticulously preparing for their respective public market debuts. Anthropic confidentially filed for an Initial Public Offering (IPO) on June 1st, with OpenAI following suit just a week later. These filings set the stage for what could potentially become two of the largest and most closely watched U.S. stock market debuts in history, reflecting the immense investor confidence and anticipated growth in the AI sector.
The Strategic Imperative of Robotics in the AI Race
The sudden intensified interest in robotics by these leading AI firms raises a critical question: why robotics, and why now? The most compelling answer lies in the evolving understanding of what constitutes true "superintelligence" or Artificial General Intelligence (AGI). Many researchers and strategists now posit that a fundamental comprehension of the physical world is an indispensable prerequisite for achieving such advanced systems. No amount of abstract internet text data, however vast, can fully substitute for the direct, empirical understanding derived from interaction with the real, tangible environment.
OpenAI’s own tumultuous history with robotics provides an instructive case study. The company initially ventured into robotics, famously developing an advanced robotic hand capable of solving a Rubik’s Cube. However, in 2021, OpenAI abruptly disbanded its entire robotics group. Co-founder Wojciech Zaremba later articulated that this early approach, while technically impressive, was missing crucial components necessary for realizing genuine superintelligence. Yet, the strategic importance of physical embodiment never truly vanished from OpenAI’s long-term vision. By 2024, the robotics initiative quietly resurfaced, with the company establishing a clandestine humanoid robotics lab in San Francisco. This clandestine effort was brought into the public eye in late May when CEO Sam Altman officially announced the relaunch of "OpenAI Robotics," actively recruiting talent and outlining a near-term focus on developing robots for critical infrastructure work, with the ambitious long-term goal of a personal robot for every individual.
Anthropic, in contrast to OpenAI’s hardware-centric approach, has not yet established a comparable physical robotics laboratory. Its engagement with robotics has primarily manifested through its internal research group, which conducts rigorous stress-testing of frontier AI capabilities, particularly for safety purposes. A notable example is "Project Fetch," launched last November, where Anthropic researchers explored the extent to which its AI model, Claude, could assist non-experts in programming a robot dog to perform various tasks. A second phase of Project Fetch, conducted in June, yielded even more striking results: a newer iteration of Claude was found to complete the same tasks approximately 20 times faster than the most efficient human-plus-Claude team from the previous year. While impressive, these are still research-oriented simulations and demonstrations.
For Anthropic, acquiring an established team with deep expertise in physical robotics, such as Physical Intelligence, would offer an unparalleled advantage. It would allow the company to circumvent potentially years of foundational research, development, and engineering work required to build a robust robotics division from scratch. This strategic shortcut could significantly accelerate Anthropic’s trajectory in bridging the gap between advanced AI models and their application in the physical world, a crucial step towards AGI.
The Intricate Web of Shared Investors and Potential Conflicts
However, any potential acquisition of Physical Intelligence by Anthropic is complicated by a highly intricate web of shared investments and strategic alliances that characterize the current AI landscape. Physical Intelligence was founded in San Francisco roughly two years ago by Lachy Groom, alongside former Google researchers and distinguished professors from Stanford and Berkeley, bringing together a powerhouse of talent and academic rigor. What makes the situation particularly complex is that Physical Intelligence’s early investor base bears a striking resemblance to that of OpenAI. Prominent venture capital firms like Khosla Ventures and Thrive Capital are significant backers of both companies. Furthermore, Founders Fund, another major investor in OpenAI, was reportedly involved in Physical Intelligence’s most recent funding round earlier this year.
The most significant complication, however, lies in the fact that OpenAI is itself an investor in Physical Intelligence. This transforms OpenAI from a peripheral competitor into a direct stakeholder in a company that its chief rival, Anthropic, was reportedly in active acquisition talks with very recently. This revelation introduces a layer of complexity and potential conflict of interest that could significantly impact any future M&A discussions.
This unprecedented situation raises several critical questions regarding corporate governance and investor rights. Did OpenAI’s early investment in Physical Intelligence come with any specific information rights, granting them access to sensitive operational or financial data? More importantly, did their investment include a "right of first refusal" clause, a protective provision sometimes negotiated by strategic investors that grants them the option to match any third-party offer for the company, or at least a right to veto a sale to a direct competitor? Such provisions are explicitly designed to prevent scenarios where a rival could acquire a strategically important portfolio company.
Given OpenAI’s existing shareholder status in Physical Intelligence, its close professional relationship with co-founder Lachy Groom, and its declared ambition to outpace Anthropic in the critical domain of robotics, it could be argued that OpenAI possesses a more compelling and legally defensible claim to Physical Intelligence than Anthropic. The possibility exists that if Physical Intelligence were indeed "in play" for acquisition, OpenAI could leverage its position to either acquire the company itself or block a sale to Anthropic. When TechCrunch reached out to OpenAI for clarification on these crucial questions, the company did not provide a response, leaving the implications of this intricate investment landscape open to speculation and further scrutiny.
The ongoing saga of Physical Intelligence and its potential suitors underscores the intense competitive pressures and strategic maneuvering within the rapidly evolving AI industry. As both Anthropic and OpenAI prepare for their monumental IPOs, their quests for dominance in areas like robotics will likely intensify, pushing the boundaries of technology and corporate strategy in unforeseen ways. The intricate web of investments, rivalries, and strategic imperatives ensures that the race for AI supremacy will continue to be one of the most compelling and closely watched narratives in the global technology landscape.
