DENVER — Anthropic — the company that makes the artificial intelligence model Claude — is in the middle of a standoff with the Pentagon over what the U.S. military can do with its technology.
Anthropic already has a contract with the federal government to use its AI for national security reasons. But recent reports say the company has taken a stand to not allow its tech to be used for fully autonomous lethal weapons or mass surveillance of U.S. citizens.
Defense Secretary Pete Hegseth has reportedly given Anthropic an ultimatum with a Friday deadline: drop those restrictions, or risk being essentially blacklisted by the government and losing its huge contract to a competitor.
This week, Denver7’s Ryan Fish sat down with Metropolitan State University Denver communications professor and AI expert Samuel Jay, who says Anthropic’s decision will not only have big implications for the company, but also the country.
“It's very much part of their identity at Anthropic, is to be the ethical AI company and a human-centered AI company,” Jay explained. “The long-term health of that company, if it were to give into these demands, is not going to be great. I mean, that could have a huge, huge blowback in terms of their reputation.”
Anthropic this week has already revised its Responsible Scaling Policy — its framework for self-imposed limits to keep its AI models ethical — vowing more changes in order to adapt and compete in the industry.
When asked why the average American should care about whether Anthropic caves to government pressure or walks away, Jay says the former would take the U.S. government in a more authoritarian direction, comparing it to how China handles AI technology.
“The government can't pick winners and losers,” Jay said. "We're a democracy. That's not supposed to be what the U.S. government does… There's supposed to be a separation of public and private, and what happens in D.C. and what happens in Silicon Valley. And I think as that line gets blurred further and further, it's something that we ought to be very, very concerned with.”
