OpenAI-Anthropic Feud Colors AI Talk

“The last time Sam Altman and Dario Amodei stood on stage together, they awkwardly tried to avoid physical contact even as other tech leaders held hands aloft for a group photo with India's prime minister.

 

They looked like pouting children on the playground -- not the chief executives of OpenAI and Anthropic, two of the hottest names in the artificial-intelligence scene. To many, the odd exchange was the physical manifestation of the growing rivalry between the companies. Both have been weighing going public this year and, in doing so, are fighting each other for users, talent and investor dollars.

 

That dust-up was a couple of weeks ago. Things have only grown more heated as the men and their companies have tried to claim the moral high ground in conflicting dealings with the Pentagon.

 

American business and Silicon Valley, in particular, are littered with classic beefs fueled by ambition, greed and green-eyed jealousy: the late Steve Jobs vs. Bill Gates; Apple vs. Samsung; Uber vs. Lyft. Then there's Elon Musk vs., well, everyone -- Jeff Bezos, Mark Zuckerberg, even Altman.

 

Arguably such spirited rivalries can be good for consumers in the long run. Each side is trying to outinnovate, outprice, outcompete the other. But the OpenAI-Anthropic feud carries a unique risk with the fate of the still-nascent AI technology resting in just a few powerful hands.

 

The toxic turn between OpenAI and Anthropic fuels distrust and threatens to further rupture consensus around still-evolving safety practices. It will color public debate around how the technology should be used and influenced.

 

Heightened tensions exploded into view in recent days as each company tried to navigate its own relationship with the Pentagon and its demands for control over AI. The messy dance saw Anthropic lose its business with the government while OpenAI gained new ground. The raw bitterness displayed between the rivals underscores the challenge the companies will have working together.

 

"This is not the last time we will see state interference into frontier AI, and until we build formalized structures for such interference it will be important for the industry to hang tough together," Dean Ball, a former Trump administration AI adviser, posted on X. "I fear that will be less likely now."

 

Anthropic was famously born in 2021 out of safety concerns that Amodei -- then OpenAI's vice president of research -- had about his employer's approach to AI.

 

Whereas Altman can be seen as a wheeler-dealer, racing to cut big deals to grow fast, Amodei staked out a comparably measured position -- almost academic or, some might say, zealous. He is known for voicing concerns around the potential dark side of the power he is developing, warning of Great Depression-like job losses.

 

To be fair, Altman, too, voices concerns, but in an almost gee-whiz wonderment that leaves no doubt he's on the side of the robot.

 

Not surprisingly, the two have been on opposite sides of policy debates around regulation. For example, in California, Anthropic backed a first-of-its kind law aimed at providing some sort of guardrails. The measure was widely decried by many in tech, who warned it would unleash patchwork regulation that would stymie AI development.

 

As this year began and it became clear both companies were thinking about IPOs, Amodei started taking some veiled shots at OpenAI. This appeared to be an effort to draw a finer line between the two.

 

Amodei, in January at Davos, essentially questioned the ethics of AI companies run by leaders who came out of social media. That was an unnamed swipe at Altman, who dropped out of Stanford University to co-found a social-network startup and later became a big backer of Reddit.

 

In early February, Anthropic launched a marketing campaign, including Super Bowl commercials, which took aim at OpenAI's plans, without specifically naming the company, to bring ads to its chatbot.

 

Altman responded that the ads were dishonest. "I guess it's on brand for Anthropic doublespeak to use a deceptive ad to critique theoretical deceptive ads that aren't real," Altman posted on X.

 

It felt as if Anthropic was on the ascent, threatening to outshine OpenAI.

 

By month's end, the stakes between the companies had grown even bigger when Amodei rejected Pentagon efforts to get Anthropic to throw out its redlines on using its technology for mass domestic surveillance and fully autonomous weapons.

 

The principled stand threw Anthropic's business into jeopardy. The federal government not only banned its use, but Defense Secretary Pete Hegseth labeled the company a supply-chain risk. Such a designation threatens Anthropic's ability to transact with companies doing business with the U.S. government.

 

Then Altman swooped in to cut his own deal with the Pentagon about the same time he went public in apparent support of Anthropic's position. While Altman characterized his arrangement as safeguarding OpenAI's own similar redlines, the move was seen as, at best, opportunistic.

 

In the wake of that, Amodei dashed off an emotional note to his staff that took aim at OpenAI and its claims ("the mendacious nature of it" as he began in the message reported by The Information, the tech publication.)

 

"Mendacious" is a five-dollar word for lying.

 

"We haven't given dictator-style praise to Trump (while Sam has)," Amodei continued. "We have supported AI regulation which is against their agenda, we've told the truth about a number of AI policy issues (like job displacement), and we've actually held our red lines with integrity rather than colluding with them to produce 'safety theater' for the benefit of employees. . . ."

 

After the message became public last week, Amodei apologized for the memo's tone and said his thinking had changed.

 

Unsurprisingly, the unvarnished thoughts couldn't be put back into the bottle. In return, Altman took his own swipe at his rival in a public setting. "The government is supposed to be more powerful than private companies," he said at a Morgan Stanley conference Thursday.

 

The irony is that the two men can often sound similar when they talk about AI -- its rapid acceleration and game-changing future. Both agree something big is coming. Soon.

 

Following the blowup with the Pentagon, Amodei told CBS News that Congress should weigh in on how AI could be used for mass surveillance in ways that he suggested haven't caught up with current laws. Similarly, on Thursday, Altman said elected officials should determine how AI is used in national defense.

 

Still, it's hard to imagine the two lobbying hand in hand soon.” [1]

 

1. OpenAI-Anthropic Feud Colors AI Talk. Higgins, Tim.  Wall Street Journal, Eastern edition; New York, N.Y.. 09 Mar 2026: B4.