Altman Testifies in Court: Musk's Psychological Tactics Caused 'Huge Damage' to OpenAI's Culture

OpenAI CEO Sam Altman testified in court that Elon Musk had caused "huge damage" to OpenAI's corporate culture. This explosive testimony came during Musk's lawsuit against OpenAI, revealing the tip of the iceberg regarding internal power struggles during the early days of the world's most influential AI company.

OpenAI was founded in December 2015, initially operating as a nonprofit research laboratory. The founding team included Sam Altman, Elon Musk, Greg Brockman, Ilya Sutskever, Jessica Livingston, and others. At its inception, the organization received approximately $1 billion in funding pledges, with Musk personally committing around $500 million (the actual amount disbursed remains disputed—Musk's side claims approximately $44 million). OpenAI's founding charter explicitly stated its mission to "ensure that artificial general intelligence (AGI) benefits all of humanity," with a commitment to sharing research results openly. This nonprofit-plus-open-source positioning was unique in the AI field at the time, aimed at countering the monopolization of AI technology by commercial giants like Google DeepMind. However, as deep learning's computational demands grew exponentially, the funding bottleneck of a purely nonprofit model became increasingly apparent, ultimately prompting OpenAI to create a "capped-profit" subsidiary structure in 2019—laying the groundwork for future controversies.

Altman's Court Testimony: Musk Demanded to "Take a Chainsaw" to Staff

Altman revealed during the trial that Musk had asked OpenAI President Greg Brockman and former Chief Scientist Ilya Sutskever to rank researchers by their achievements and demanded they "take a chainsaw through a bunch" of those ranked at the bottom.

This aggressive approach to personnel elimination stood in stark contrast to the academic research-oriented culture OpenAI had always tried to cultivate. As an organization whose mission was to "ensure that artificial general intelligence benefits all of humanity," OpenAI's early culture more closely resembled an academic laboratory than a high-pressure Silicon Valley startup. Musk's management style—emphasizing brutal rankings and mass layoffs—was more reminiscent of the iron-fisted approach he implemented at Tesla and Twitter (now X).

The practice Musk advocated is known in management theory as "Stack Ranking" or "Rank and Yank." This system was first popularized by former General Electric (GE) CEO Jack Welch in the 1980s, requiring managers to categorize employees into the top 20%, middle 70%, and bottom 10% by performance, with periodic elimination of the bottom group. Microsoft also widely adopted this system in the 2000s but officially abandoned it in 2013 due to its severe damage to team collaboration and innovation. In AI research, this system is particularly destructive: fundamental research outcomes often take years to materialize, making short-term rankings incapable of accurately measuring researchers' true contributions. Furthermore, forced ranking incentivizes individual competition over knowledge sharing, creating a fundamental contradiction with the highly interdisciplinary collaborative nature of AI research.

How Musk's "Mind Games" Undermined OpenAI's Organizational Culture

Altman characterized Musk's behavior as "mind games" and explicitly stated that these practices caused "huge damage" to OpenAI. While Altman made certain concessions in his testimony, his core argument was crystal clear: Musk's style of intervention destroyed the organizational atmosphere that OpenAI relied upon to attract and retain top AI talent.

For an AI research institution, the importance of culture cannot be overstated. Top-tier talent capable of frontier AI research is extremely scarce globally—by industry estimates, core researchers who can drive breakthroughs in areas like large language models and reinforcement learning number no more than a few thousand worldwide. These talents are primarily concentrated in a handful of institutions including OpenAI, Google DeepMind, Anthropic, and Meta FAIR, with exceptionally fierce talent competition among them. Top AI researchers can command annual compensation of millions or even tens of millions of dollars, but compensation is often not their primary factor in choosing an employer—research autonomy, computational resources, colleague quality, and organizational sense of mission are equally critical.

The founding of Anthropic is a classic example: in 2021, OpenAI's former VP of Safety Dario Amodei led over a dozen core researchers in a collective departure to establish Anthropic, partly due to dissatisfaction with OpenAI's internal culture and direction. This demonstrates that any negative change in organizational culture can trigger a chain reaction of talent attrition. The high-pressure ranking and layoff culture pushed by Musk could directly threaten OpenAI's core competitive advantage in attracting such talent.

The Full Story of Musk's Feud with OpenAI

Musk was one of OpenAI's co-founders and provided substantial financial support during the company's early stages. However, he departed OpenAI's board in 2018, after which relations between the two parties deteriorated rapidly.

Regarding Musk's departure, the official reason given was to avoid potential conflicts of interest with Tesla in the AI domain (particularly autonomous driving). But multiple sources indicate the reasons were far more complex. Reportedly, between late 2017 and early 2018, Musk proposed that he assume the role of OpenAI's CEO and gain control, but was rejected by Altman and the board. Additionally, Musk was dissatisfied with OpenAI's pace of development and competitive strategy, believing OpenAI was falling too far behind Google in the AI race. After leaving, Musk ceased funding OpenAI, and the resulting financial pressure became a significant driver behind OpenAI's 2019 transition to a "capped-profit" model and its acceptance of Microsoft's $1 billion investment. This transformation later became the central point of contention in Musk's lawsuit.

Musk has repeatedly and publicly criticized OpenAI for deviating from its original nonprofit mission, shifting toward a profit-driven, closed development model—especially after establishing its deep partnership with Microsoft. Musk subsequently founded his own AI company, xAI, in July 2023, headquartered in the San Francisco Bay Area, with the core mission of "understanding the true nature of the universe." xAI quickly assembled a research team drawn from top institutions including Google DeepMind, Microsoft Research, and the University of Toronto, and launched its first large language model, Grok, in November 2023, integrated into the X platform. xAI completed multiple rounds of large-scale funding in 2024, with its valuation rapidly climbing to tens of billions of dollars, and built a massive GPU cluster called "Colossus" in Memphis, once described as one of the world's largest AI training clusters. The establishment of xAI transformed Musk from a critic of OpenAI into a direct competitor, adding a commercial competition dimension to his lawsuit against OpenAI—a point OpenAI has repeatedly raised, arguing that Musk's litigation motives are not purely rooted in public interest considerations.

Altman's court testimony now reveals management controversies during Musk's tenure from OpenAI's internal perspective, adding a new dimension to this protracted legal and public relations battle.

Deeper Implications of This Lawsuit for the AI Industry

This dispute is far more than a personal feud between two tech titans—it reflects several fundamental issues facing the AI industry:

The philosophical battle over AI talent management. In a field as heavily dependent on top talent as AI, should companies adopt Musk's high-pressure elimination system, or cultivate a more inclusive research environment? Different choices directly impact an organization's innovative capacity and long-term competitiveness. Historical experience shows that research institutions that produced epoch-making innovations—Bell Labs, Xerox PARC, and others—were all renowned for their relaxed and free research cultures, providing an important reference for AI industry talent management.

The boundaries of founder influence. Even co-founders and primary funders should have reasonable limits on their intervention in organizational culture. Musk's case demonstrates that excessive imposition of personal will can have destructive effects on an organization.

The complexity of AI company governance. OpenAI's transition from nonprofit to for-profit, its board crisis, and founder lawsuits—this series of events highlights the enormous challenge AI companies face in balancing mission-driven goals with commercial interests. The November 2023 board crisis was particularly emblematic: OpenAI's nonprofit board suddenly announced Altman's removal as CEO, citing his lack of candor in communications with the board. This decision triggered massive internal upheaval—over 700 employees (approximately 95% of the total) signed a joint letter threatening mass resignation, and Microsoft swiftly expressed willingness to take in Altman and his team. Just five days later, Altman was reinstated as CEO, and the board underwent large-scale restructuring. The deeper causes of this "palace coup" involved a philosophical clash between the AI safety faction and the commercialization faction. Chief Scientist Ilya Sutskever was believed to be one of the key figures driving the removal; he later left OpenAI and founded Safe Superintelligence Inc. (SSI), a new company focused on AI superalignment. This series of events fully exposed the deep tension within AI companies between mission-driven governance and commercial operations.

Conclusion

As the Musk vs. OpenAI case continues to unfold, more details about OpenAI's early internal operations may come to light. Regardless of the legal outcome, this dispute has already sounded an alarm for the entire AI industry: while pursuing technological breakthroughs, building a healthy organizational culture and governance structure may be the key factor determining the long-term success or failure of AI companies.