“Exploring the Buzz: The Anthropic-Physical Intelligence Controversy Ignite AI Twitter”

This year has seen significant activity in AI acquisitions, with companies like Anthropic and OpenAI aggressively acquiring various tech startups. Recently, a rumor surfaced about Anthropic’s interest in acquiring the robotics company Physical Intelligence, which spread quickly even after being publicly denied by its CEO.

Physical Intelligence stands out in the robotics field due to its co-founder Lachy Groom, a notable figure in Silicon Valley, and its considerable funding exceeding $1 billion. The company is known for its π0.5 model, recognized as one of the leading robotics systems used in research.

According to reports, the rumor held some truth as Anthropic and Physical Intelligence did engage in acquisition discussions earlier this year. While tech blogger Robert Scoble’s social media post fueled the speculation, it seems he may have misrepresented the particulars while still being correct about the interactions occurring.

The response from Physical Intelligence was somewhat ambiguous. CEO Karol Hausman reportedly dismissed the rumors in a Slack message featuring a gif of a character shaking her head, signaling uncertainty rather than a firm denial.

Groom did not respond to requests for comment on this matter, leaving some questions unanswered.

This year, Anthropic has confirmed four acquisitions, while OpenAI has taken a more aggressive stance, securing at least 17 firms in 2023. Both companies are now preparing for public offerings, with Anthropic submitting its IPO request on June 1, followed by OpenAI a week later, potentially leading to two of the largest IPOs in U.S. history.

The growing interest in robotics stems from the belief that tangible world understanding is essential for developing advanced AI systems, which cannot be entirely achieved through virtual data alone.

OpenAI’s past involvement with robotics provides context. Initially, it created a robotic hand capable of solving a Rubik’s Cube but discontinued its robotics division in 2021 due to gaps in its approach toward achieving superintelligence. In 2024, the company re-entered the field, establishing a new robotics lab with plans aimed at creating personal robots and enhancing infrastructure.

In contrast, Anthropic has focused on publishing research to enhance AI safety, including a project last November where they evaluated how their model could assist in programming a robotic dog, achieving significant efficiency improvements in a later phase this June.

Acquiring Physical Intelligence would enable Anthropic to bypass lengthy development periods by integrating an already-established team with valuable expertise. However, challenges exist; Physical Intelligence has strong connections to influential investors similar to those of OpenAI, raising questions about potential conflicts or preemptive rights related to ownership changes.

Furthermore, OpenAI’s previous investment in Physical Intelligence complicates matters, as it could imply a vested interest that might provide it with an upper hand if the company is indeed being considered for acquisition. OpenAI has not commented on these inquiries.