.png)
In Brief
A major Wall Street Journal investigation traces the OpenAI-Anthropic rivalry back to a San Francisco group house in 2016, and revealed it's driven as much by personal grudges and power struggles as AI philosophy and safety.
The competitive fallout is reshaping both companies in real time, including OpenAI shutting down its Sora video app to refocus its resources.
What Happened
The Wall Street Journal published an in-depth investigation into the personal history behind the OpenAI-Anthropic rivalry, based on interviews with current and former employees at both companies. Many of the details had never been reported.
The tensions began in 2016 at a San Francisco group house where Dario Amodei, his sister Daniela, and her fiance Holden Karnofsky (an effective altruism philanthropist) lived. Greg Brockman, OpenAI co-founder, frequented the house and debated AI philosophy with them. Early disagreements over whether to inform the public or the government first about AI's trajectory deepened when Brockman floated the idea of selling AGI to the UN Security Council's nuclear powers. Dario considered the proposal "tantamount to treason."
After Elon Musk exited in 2018 and Sam Altman took over OpenAI, internal conflicts escalated. Dario blocked Brockman from working on the GPT language model project. Daniela, who was co-leading it, offered to step down rather than let Brockman join. By 2020, Altman had accused the Amodeis of plotting against him to the board. Dario, Daniela, and nearly a dozen employees left to found Anthropic, and Dario wrote a memo arguing the ideal AI company would be 75% public good and 25% market. Five years later, both companies are valued at hundreds of billions of dollars and racing toward IPOs.
On Episode 207 of The Artificial Intelligence Show, SmarterX founder and CEO Paul Roetzer unpacked why these personal dynamics matter for every business using AI.
The Key Numbers
- 2016 - The year the tensions began between key AI leaders who then were working for OpenAI
- 2020 - The year the rivalry officially split the company, with Dario Amodei, his sister Daniela, and nearly a dozen other employees leaving OpenAI to found Anthropic.
- 75% and 25% - The ideal business structure Dario Amodei proposed in a memo before his departure, arguing that an AI company should be 75% for the public good and 25% market.
- $25 million - Amount of a recent donation by OpenAI's Greg Brockman to a pro-Trump Super PAC
Why Knowing About the AI Leaders Matters
The rivalries have real commercial consequences. In recent months, Anthropic CEO Dario Amodei has compared the Altman-Musk legal battle to "Hitler and Stalin," called Brockman's $25 million pro-Trump Super PAC donation "evil," and likened OpenAI to a tobacco company. Internally, Anthropic refers to this brand strategy as positioning itself as the "healthy alternative." On the other side, OpenAI shut down its Sora video app this week, which was burning $1 million per day with fewer than 500,000 users. This was done in part to redirect compute resources to compete with Anthropic's enterprise gains.
"Every day I'm talking to companies and leaders of companies who are moving to Anthropic," Roetzer says. "It is a common recurring theme I'm hearing."
Brockman is the recurring friction point. The WSJ piece and Roetzer's analysis of past friction reveal a consistent pattern: Brockman inserting himself into projects and creating internal tension. This played out with the GPT project, the meeting with President Barack Obama that Dario was excluded from, and ultimately Brockman's 2024 sabbatical that the Journal later revealed was a mutual agreement between Brockman and Altman due to management friction. "Over and over and over again, you hear this throughout these issues through the years," Roetzer says.
"A lot of this does stem from Dario Amodei not getting the kind of credit he thought he deserved for his contributions to really the whole transformational phase we find ourselves in with language models."
Paul Roetzer, founder and CEO of SmarterX, Episode 207 of The Artificial Intelligence Show
Politics makes this volatile. Roetzer laid out the political positioning of the five frontier AI labs. Google DeepMind has stayed largely neutral. OpenAI was historically neutral but is now tied to the current administration through Brockman's Super PAC donation. Anthropic is effectively the administration's adversary, fighting the Pentagon. Meta and xAI are fully aligned with the current political direction. "What happens if in two years, if the power switches, and then the companies that have gone all in on one side or the other, what happens to those labs?" Roetzer asks.
Five companies will shape nearly everything. The three tier-one labs (Google DeepMind, OpenAI, Anthropic) plus Meta and xAI control the frontier of AI development. They need funding, data centers, energy infrastructure, compute, and the most powerful models. Two of the top three are at war with each other. A third tier-two lab is suing one of them. "These five companies are going to decide everything when it comes to the economy, business, geopolitics," Roetzer says.
SmarterX Take
The Sora shutdown is the most concrete signal of how this rivalry is redirecting resources. Fidji Simo, OpenAI's head of applications, described Anthropic's enterprise market gains as a "wake-up call" and told staff the company "cannot miss this moment because we are distracted by side projects." When a company kills a product that Disney was about to invest $1 billion in, the competitive pressure is intense.
For business executives choosing which AI platforms to build on, understanding these dynamics is essential because of. how influential their leaders are. Which lab wins the enterprise race, which one has the best political positioning, and which ones remain viable through the next political cycle are all factors that affect long-term technology decisions. And in reality, the personal grudges driving these companies are shaping the products, partnerships, and priorities that directly affect every organization using AI.
What to Watch
Both companies are racing toward IPOs this year. Anthropic and OpenAI are competing not just for enterprise customers but for public market positioning. How each company tells its story to investors will reveal what they actually prioritize when the money is on the line.
The Musk-Altman trial is imminent. The lawsuit between Elon Musk and Sam Altman goes to trial soon. The outcome could reshape OpenAI's corporate structure and competitive position, adding yet another layer of instability to the industry's most consequential rivalry.
Further Reading
The Decadelong Feud Shaping the Future of AI → wsj.com
Why OpenAI Really Shut Down Sora → techcrunch.com
Heard on The Artificial Intelligence Show, Episode 207
Paul Roetzer and Mike Kaput discuss the personal history behind the OpenAI-Anthropic rivalry and why it matters for every business making AI decisions for their companies. Listen →
Mike Kaput
Mike Kaput is the Chief Content Officer at SmarterX and a leading voice on the application of AI in business. He is the co-author of Marketing Artificial Intelligence and co-host of The Artificial Intelligence Show podcast.