According to recently unsealed legal filings, the artificial intelligence research organization Anthropic engaged in preliminary discussions about a potential merger with its notable rival, OpenAI, in the immediate aftermath of the brief and dramatic removal of Sam Altman as OpenAI’s chief executive officer two years ago. These documents reveal that the idea of uniting two of the industry’s most influential labs emerged during a particularly turbulent period in OpenAI’s leadership history.

The details of these extraordinary merger talks were brought to light through testimony provided by Ilya Sutskever, a cofounder of OpenAI and its former chief scientist, during a formal deposition connected to Elon Musk’s ongoing lawsuit against Altman and OpenAI. The suit centers on the company’s alleged deviation from its original nonprofit structure and mission. During the lengthy legal session, Sutskever meticulously described the short-lived negotiations between OpenAI and Anthropic, offering new insight into the internal dynamics of the AI industry’s most prominent players.

During the ten-hour deposition, which took place in San Francisco in the previous month, Sutskever recounted that a proposal surfaced in November 2023 suggesting that OpenAI—the organization best known as the creator of ChatGPT—might combine forces with Anthropic. Had those discussions advanced, the contemplated structure would have positioned Anthropic to assume control of OpenAI’s leadership apparatus, effectively changing the balance of power between the two companies. Sutskever clarified that he found this idea deeply troubling and personally opposed, expressing discontent over the notion of relinquishing OpenAI’s leadership to another firm.

When contacted by Business Insider seeking clarification or comment regarding these developments, representatives from both OpenAI and Anthropic reportedly did not respond promptly. Similarly, Sutskever’s legal counsel declined to issue a statement, maintaining discretion in light of the ongoing legal proceedings.

During his sworn testimony, Sutskever elaborated that, immediately after the OpenAI board made the controversial decision to terminate Altman, a conversation unfolded between members of the board and the senior leadership of Anthropic—among them, CEO Dario Amodei and his sister, Daniela Amodei, who serves as the company’s president. As recounted by Sutskever, Anthropic’s representatives appeared enthusiastic about the strategic possibility, even as they acknowledged significant practical barriers that would complicate such a merger. “I recall Anthropic expressing their excitement about it and also raising the issue of the practical challenges that they would face in implementing it,” Sutskever stated, according to the court records examined by Business Insider.

At the time, Sutskever not only served as OpenAI’s chief scientist but also sat on its board of directors. He recalled being “very unhappy” about the proposal, emphasizing his firm opposition to any merger with Anthropic. However, he noted that his perspective represented a minority stance within the boardroom. Other directors were considerably more open to the concept, he explained, naming former board member Helen Toner as one of the idea’s strongest advocates. “They were a lot more supportive, yes,” Sutskever conceded, adding that, to his knowledge, none of the other members explicitly objected to the proposal.

The envisioned collaboration, however, never matured into a formal plan. Sutskever characterized the entire episode as fleeting—an “extremely brief” exchange that ultimately dissolved when certain logistical and operational difficulties surfaced. “My recollection is that there were some practical obstacles that Anthropic raised, and so the proposal did not continue,” he recounted. Under questioning from OpenAI’s attorneys, Sutskever further admitted that he was not privy to the specific details of those obstacles, implying that the discussions concluded before reaching any advanced stage of negotiation.

In the days that followed those chaotic events, Sam Altman—who had originally founded OpenAI in 2015 alongside Sutskever and Elon Musk—was reinstated as the company’s CEO after widespread employee backlash and industry attention brought pressure on the organization’s board to reverse its decision. During the October 1st deposition, lawyers representing both OpenAI and Musk alternated in questioning Sutskever about his recollections of these events and his understanding of the leadership turmoil.

The court has since ordered Sutskever to return for a second deposition at the request of Musk’s legal team. This forthcoming session is expected to delve more deeply into Sutskever’s financial interests in OpenAI and into the contents of a confidential internal memorandum—referred to in filings as the “Brockman memo”—that was allegedly authored by Greg Brockman, another OpenAI cofounder. In that document, Brockman is said to have detailed the circumstances surrounding Altman’s temporary dismissal.

The broader legal battle between Musk and Altman remains active and acrimonious. Musk filed suit against Altman and OpenAI in federal court in California last year, accusing them of straying from OpenAI’s founding mission as a nonprofit research organization intended to promote the safe and open dissemination of artificial intelligence technology for the collective benefit of humanity. The complaint alleges violations of antitrust law by both OpenAI and its strategic partner, Microsoft, asserting that they pressured investors not to support competing ventures such as Musk’s own AI company, xAI. It further claims that Altman and other company executives misled Musk during the formation of OpenAI, persuading him to participate under false pretenses regarding the enterprise’s nonprofit objectives.

In response, OpenAI filed a countersuit earlier this year, accusing Musk of engaging in a prolonged campaign of intimidation and reputational attacks against the organization and its leadership. The enmity between Musk and Altman has spilled far beyond court documents, extending into highly publicized social media exchanges that continue to captivate the technology community. Over the recent weekend, the two billionaires traded charged remarks on X, Musk’s own social media platform. Musk publicly accused Altman of effectively stealing control of a nonprofit organization, writing, “You stole a non-profit,” in reply to an unrelated post by Altman concerning a Tesla Roadster order. Altman retorted that he had “helped turn the thing [Musk] left for dead into what should be the largest nonprofit ever,” adding that the current corporate structure of OpenAI was a necessary prerequisite for achieving such broad-scale ambitions.

Only weeks prior, OpenAI had officially announced the completion of its conversion into a for-profit public benefit corporation—a hybrid structure designed to balance commercial growth with a commitment to public good. That transformation underscores the broader philosophical debate at the heart of this dispute: whether artificial intelligence, as a transformative technology, can simultaneously thrive in a commercial marketplace while remaining faithful to the ethical imperatives of safety, accessibility, and humanity’s well-being.

Sourse: https://www.businessinsider.com/openai-cofounder-ilya-sutskever-anthropic-merger-discussions-deposition-2025-11