From shared housing disputes to a 300 billion showdown: The Wall Street Journal's in-depth article reveals for the first time the ten-year private feud between Anthropic and OpenAI founders

Title: “The Decadelong Feud Shaping the Future of AI”

Author: Keach Hagey

Source:

Reprinted from: Mars Finance

Wall Street Journal reporter Keach Hagey published an in-depth investigative report, systematically revealing for the first time the decade-long personal feud between the founders of Anthropic and OpenAI through extensive interviews with current and former employees and executives of both companies. What influences the global AI landscape is not only a rivalry over technological approaches but also an unhealed personal trauma.

In recent months, Dario Amodei’s internal language has been much more intense than in public. He compared the legal disputes between Sam Altman and Elon Musk to “the conflict between Hitler and Stalin,” calling OpenAI President Greg Brockman’s $25 million donation to a pro-Trump super PAC “evil,” and likening OpenAI and other competitors to “tobacco companies that knowingly sell harmful products.”

After the dispute with the Pentagon escalated, he also called OpenAI “mendacious” on Slack, writing, “These facts reveal a behavioral pattern I often see in Sam Altman.”

Internally, Anthropic describes this branding strategy as creating a “healthy alternative” to competitors. During this year’s Super Bowl, a satirical ad that indirectly mocked OpenAI for embedding ads in its chatbot was a public product of this strategy.

The story begins in 2016 in the living room of a shared house on Delano Street in San Francisco. Dario and his sister Daniela Amodei lived there, and Brockman, co-founder of OpenAI, often visited due to his close friendship with Daniela. One day, Brockman, Dario, and Daniela’s then-fiancé, effective altruism philanthropist Holden Karnofsky, debated the correct development path for AI: Brockman believed all Americans should be informed about the latest AI developments, while Dario and Karnofsky argued that sensitive information should first be reported to the government rather than broadcast publicly. This disagreement later became a philosophical dividing line for both companies.

Impressed by OpenAI’s talent pool, Dario joined in mid-2016 and stayed up late training AI agents to play video games with Brockman. However, after four years working together, tensions over power and a sense of belonging deepened. In 2017, Musk, then the main investor in OpenAI, demanded a list of each employee’s contributions to justify layoffs. About 10% to 20% of the team of around 60 people was dismissed one by one, which Dario saw as cruel; one of those laid off later became a co-founder of Anthropic.

That same year, an ethics advisor hired by Dario proposed that OpenAI serve as a coordinating entity between AI companies and the government. Brockman extrapolated this to the idea of “selling AGI to the UN Security Council’s nuclear powers,” which Dario considered nearly treasonous and briefly contemplated resigning.

After Musk’s departure in 2018, Altman took over leadership. He and Dario reached a consensus: employees lacked confidence in Brockman and Chief Scientist Ilya Sutskever’s leadership. Dario agreed to stay on the condition that they would no longer be in charge, but soon discovered that Altman had simultaneously promised the latter two that they could fire him, creating a contradiction.

As the development of the GPT series began, conflicts among executives over who could participate in the language model project intensified. Dario, then research director, refused to allow Brockman to interfere. Daniela, co-leading the project with Alec Radford, threatened to resign as head to exert pressure. Radford’s personal wishes became entangled in the proxy war among executives.

Dario’s credentials rose with the success of GPT-2 and GPT-3, but he felt Altman downplayed his contributions. When Brockman discussed the OpenAI charter on a podcast, Dario was angry that he had contributed more but was not invited to participate. He was equally displeased when he learned Brockman and Altman planned to meet with former President Obama without including him.

The conflict reached a boiling point during a confrontation in a meeting room. Altman summoned the Amodei siblings and accused them of inciting colleagues to submit negative feedback about him to the board. Both denied it. Altman claimed the information came from another executive, and Daniela immediately called that executive in to confront them, who said they were completely unaware.

Altman then denied having said such a thing, leading to a heated argument. In early 2020, Altman asked executives to write peer reviews of each other. Brockman wrote a strongly worded critique accusing Daniela of abusing her power and using bureaucratic procedures to exclude dissenters. Altman reviewed it in advance, describing it as “tough but fair.” Daniela rebutted each point, and the argument escalated to the point where Brockman once suggested retracting his comments.

By late 2020, a team led by Dario decided to leave. Daniela led negotiations with lawyers regarding their departure. Altman personally visited Dario’s home to persuade him to stay, but Dario insisted he would only accept reporting directly to the board and made it clear he could not work with Brockman. Before leaving, he wrote a lengthy memo dividing AI companies into “market-oriented” and “public-interest-oriented” categories, believing the ideal ratio was 75% public interest and 25% market. A few weeks later, Dario, Daniela, and nearly a dozen employees left OpenAI to establish Anthropic.

Today, five years later, both companies are valued at over $300 billion and are competing fiercely for IPOs. During the closing photo at the AI summit in New Delhi this February, Indian Prime Minister Modi and other tech leaders raised their hands, while Amodei and Altman chose not to participate, only awkwardly bumping elbows.