As Sam Altman discovered Saturday night, these are tough times to work in the U.S. government. Around 7 p.m., OpenAI’s CEO announced that he would be holding public inquiries regarding X as a way to shed light on his company’s decision to take over the Department of Defense contract that Anthropic had just pulled out of.
Most of the questions centered around OpenAI’s willingness to engage in mass surveillance and automated killing, the very activities Anthropic had ruled out in its negotiations with the Pentagon. Mr. Altman typically spoke out to the public sector, saying it was not his role to set national policy.
“I believe very deeply in the democratic process,” he wrote in a reply, “and that elected leaders have power and that we all must uphold the Constitution.”
An hour later, he confessed that he was surprised that so many people seemed to be against it. “There’s a more open discussion than I thought about whether democratically elected governments or unelected private corporations should have more power, and I think people are probably divided on this,” Altman said.
This is an important moment for both OpenAI and the technology industry as a whole. During the Q&A, Altman took a stance that is standard in the defense industry, where military leaders and industry partners are expected to follow civilian guidance.
But more importantly, as OpenAI transitions from a highly successful consumer startup to part of the national security infrastructure, the company appears ill-equipped to manage its new responsibilities.
Mr. Altman’s City Hall was founded at the best time for his company. The Pentagon had just blacklisted OpenAI rival Anthropic for asserting contractual restrictions on surveillance and autonomous weapons. A few days later, OpenAI announced that it had won the same contract that Anthropic passed up. Altman described the deal as a quick way to defuse the conflict, and it was certainly a lucrative one. But he seemed unprepared for how much of a backlash it caused from both the company’s users and employees.
tech crunch event
San Francisco, California
|
October 13-15, 2026
OpenAI has been working with the U.S. government for years, but never like this. For example, when Mr. Altman was arguing before a Congressional committee in 2023, he was still following a social media strategy. He touted the company’s world-changing potential, recognizing the risks and engaging passionately with lawmakers. This is the perfect combination to stir up investors while circumventing regulation.
Less than three years later, that approach was no longer acceptable. It is clear that AI is so powerful and the demand for funding is so intense that more serious engagement with governments is impossible to avoid. What is surprising is how unprepared both sides seem for it.
The biggest immediate conflict is between Anthropic itself and Secretary of Defense Pete Hegseth’s announced plan on Friday to designate the institute as a supply chain risk. That threat looms over the entire conversation like an unfired gun. As former Trump official Dean Ball wrote over the weekend, this designation would cut Anthropic off from its hardware and hosting partners, effectively putting it out of business. This is an unprecedented action against a U.S. company, and while it may ultimately be reversed in court, it will cause damage and shock the industry in the meantime.
As Mr. Ball explained the process, Anthropic was fulfilling an existing contract based on terms established years ago, only for the government to request a change to the terms. This is much more than what’s being thrown around among private companies, and it sends a chilling message to other vendors.
“Even if Secretary Hegseth stands back and narrows down the very broad threat to humanity, significant damage has been done,” Ball wrote. “Most businesses, political activists, and other businesses will have to operate under the assumption that tribal logic will prevail from now on.”
While this is a direct threat to Anthropic, it is also a serious problem for OpenAI. The company is already under intense pressure from its employees to maintain certain red lines. At the same time, right-wing media will be wary of any signs that OpenAI is not such a strong political ally. In the middle of it all is the Trump administration, doing its best to make the situation as difficult as possible.
You could argue that OpenAI was never intended to be a defense contractor, but its massive ambitions force it to play the same game as Palantir and Anduril. Moving forward during the Trump administration means choosing a side. There are no apolitical people here, and gaining some friends means alienating others. It remains to be seen how much OpenAI will pay in terms of lost business and lost employees, but it’s unlikely to come out unscathed.
It may seem strange that this crackdown comes at a time when more prominent tech investors than ever are occupying influential positions in Washington, but most of them seem perfectly comfortable with the tribe’s logic. Anthropic has long been perceived among pro-Trump venture capitalists as favoring the Biden administration to the detriment of the larger industry, a perception underscored by Trump adviser David Sachs’ response to the ongoing conflict. Now that that reversal has occurred, few seem willing to support the broader principle of free enterprise.
This is a difficult position for any company. Politically aligned companies may benefit in the short term, but they are equally at risk when political winds inevitably shift. There’s a reason the defense sector has been dominated for decades by slow-moving, heavily regulated conglomerates like Raytheon and Lockheed Martin. Operating as the industrial arm of the Pentagon gave it the political cover it needed to avoid politics and allowed it to stay focused on technology without having to hit the reset button every time the White House changed hands.
Today’s startup competitors may move faster than their predecessors, but they are far less prepared for the long term.
