A personal conflict between Sam Altman and Dario Amodei at OpenAI sparked a philosophical divide that ultimately created one of the world's most valuable AI companies. What began as disagreements over how to develop artificial general intelligence (AGI), a hypothetical AI system that could match or exceed human intelligence across all domains, evolved into internal power struggles that culminated in Amodei's departure and the founding of Anthropic in late 2020. Today, both companies are valued at over $300 billion and racing toward public offerings, yet the tension between their leaders remains palpable. What Started the OpenAI Conflict? The roots of the conflict trace back to 2016 in a shared house on Delano Street in San Francisco. Dario Amodei, his sister Daniela, and OpenAI co-founder Greg Brockman engaged in a pivotal debate about the right path for AI development. Brockman believed all Americans should be informed about advances at the frontier of AI research, while Dario and effective altruism philanthropist Holden Karnofsky argued that sensitive information should first be reported to the government before public disclosure. This fundamental disagreement would later define the philosophical paths of both companies. Impressed by OpenAI's team, Dario joined the organization in mid-2016. For four years, he worked closely with Brockman, staying late into the night to train AI agents to play video games. However, conflicts over power and organizational belonging deepened steadily. In 2017, when Elon Musk was OpenAI's main funder, he demanded a list of each employee's contributions and used it as a basis for layoffs, with 10 to 20 percent of the 60-person team being terminated one by one. Dario viewed this approach as cruel, and one of those laid off would later become a co-founder of Anthropic. How Did Power Struggles Escalate at OpenAI? The situation intensified after Altman took over leadership in 2018 following Musk's departure. Dario and Altman initially reached a consensus that employees lacked confidence in Brockman and Chief Scientist Ilya Sutskever's leadership. Dario agreed to stay only on the condition that the two no longer directly oversee him, but he soon discovered that Altman had simultaneously promised Brockman and Sutskever the right to fire him, creating conflicting commitments. This pattern of contradictory promises would become emblematic of the trust breakdown. The most intense conflict erupted over who could participate in the language model project that would eventually become the GPT series. Dario, then the research director, did not allow Brockman to intervene in the project. Daniela, who co-led the project with Alec Radford, threatened to resign from her leadership position, dragging personal preferences into what had become an executive proxy war. When Dario's credentials soared following the success of GPT-2 and GPT-3, he felt Altman downplayed his contributions. He was particularly upset when Brockman appeared on a podcast discussing OpenAI's charter without inviting him, despite his greater contribution to the document. The conflict reached a breaking point during a confrontational meeting in a conference room. Altman summoned the Amodei siblings and accused them of inciting colleagues to submit negative feedback about him to the board. The siblings denied the allegation. When Altman claimed the information came from another executive, Daniela immediately brought that executive in for questioning, who denied having any knowledge of such conversations. Altman then denied ever making the accusation, leading to a heated argument that exposed the depth of mistrust. Steps to Understanding the Philosophical Divide Between OpenAI and Anthropic - Transparency Philosophy: OpenAI under Altman pursued a more public-facing approach to AI development, believing in broad disclosure of advances to the American public and the world. - Cautious Disclosure: Anthropic, founded by Amodei, prioritizes reporting sensitive AI research findings to government entities before public announcement, reflecting a more measured approach to AGI development. - Organizational Structure: Amodei proposed dividing AI companies into "market-driven" and "public-benefit-driven" categories, believing the ideal ratio should be 75 percent public benefit and 25 percent market-focused work. - Safety Focus: Anthropic internally refers to its brand strategy as creating a "healthy alternative" to competitors, with public messaging emphasizing responsible AI development compared to what it views as riskier approaches. By early 2020, peer reviews conducted at Altman's request further damaged relationships. Brockman wrote a strongly worded feedback accusing Daniela of power abuse and using bureaucratic processes to exclude dissent, which Altman pre-approved as "tough but fair". Daniela countered each point, escalating the argument to such a degree that Brockman briefly suggested retracting his comments. Why Did Amodei Leave to Found Anthropic? By the end of 2020, the team centered around Dario decided to leave OpenAI. Daniela took the lead in negotiating departure matters with lawyers. Altman personally visited Dario's home to persuade him to stay, but Dario insisted on only reporting directly to the board and made clear he could not work with Brockman. Before departing, Dario wrote a lengthy memorandum outlining his vision for AI company structure, dividing organizations into "market-driven" and "public-benefit-driven" categories, with his ideal ratio being 75 percent public benefit and 25 percent market. Several weeks later, Dario, Daniela, and nearly a dozen other employees left OpenAI to found Anthropic. The new company was built on the philosophical principles that had been rejected at OpenAI, emphasizing safety, responsible development, and government coordination on sensitive AI research. Five years after its founding, Anthropic has grown to rival OpenAI in valuation and capability, with both companies now valued at over $300 billion. The personal animosity between the two leaders remains evident even in public settings. During a group photo at the February AI Summit in New Delhi, Indian Prime Minister Modi raised his hands high with the presence of tech leaders, while Amodei and Altman chose not to participate, awkwardly opting for an elbow bump instead. This moment symbolized the enduring distance between the two men who shaped the modern AI landscape through their conflict. Internally, Dario's language about OpenAI has become increasingly intense in recent months. He has likened Altman and Elon Musk's legal disputes to a "Hitler vs. Stalin" conflict, described OpenAI President Greg Brockman's $25 million donation to a pro-Trump super PAC as "evil," and compared OpenAI and other competitors to "tobacco companies selling products they know are harmful". On Slack, he referred to OpenAI as "mendacious," writing, "These facts suggest a pattern of behavior that I have seen all too often in Sam Altman." The feud has manifested in public messaging as well. Anthropic ran a subtly mocking advertisement during the Super Bowl this year, targeting OpenAI's inclusion of advertisements in chatbots, making the competitive tension visible to mainstream audiences. What began as a personal disagreement about how to develop AGI responsibly has become a defining force in the AI industry, with two companies now competing to shape humanity's relationship with artificial intelligence.
