- Sam Altman urged the government and Anthropic to reduce tensions and work together on AI governance.
- He argued that governments should have power over AI and national security decisions.
- He said he still mostly trusts the government, although he accepts that many do not.
Relations between Anthropic and the US government have become an unusually flammable flashpoint in the broader fight over AI regulations and control. The escalation of the fight began when negotiations with the Pentagon over how Anthropic’s Claude AI model could be used collapsed over the company’s refusal to remove safeguards against fully autonomous weapons or mass domestic surveillance.
Washington’s responses, including an executive directive banning federal agencies from using Anthropic’s technology and labeling the company a “supply chain risk,” led to lawsuits over alleged constitutional violations, and a federal judge has since temporarily blocked the Pentagon’s actions.
OpenAI CEO Sam Altman apparently sees harmony as necessary on both ends of the argument.
Article continues below.
“Find a way to work together… like stop, stop things on both sides, stop the escalation on both sides and find a way to work together,” Altman said in an interview with Laurie Segall.
The security demands of AI
AI companies have exaggerated the technology’s potential in areas such as national security, even as they push for a light regulatory touch. Altman has apparently come to the conclusion that companies can’t have it both ways. If AI has as many geopolitical consequences as everyone keeps insisting, then governments will want to have a hand on the wheel.
“I don’t think it will work for our industry to say, ‘Hey, this is the most powerful technology humanity has ever built,'” Altman said. “It will be the most important part of geopolitics. It will be the largest cyber weapon the world has ever built. It will be, you know, the determining factor of future wars and protection. And we’re not going to give it to you.”
Of course, whether people are comfortable with the government controlling such an important technology is another question. Altman said he still relies primarily on the system of checks and balances, although he acknowledged that many people currently “don’t really trust the government to follow the law.”
It is a position that stands out compared to that of some AI leaders who are more distrustful of the government. However, he believes it would be a mistake not to help the government with national security, especially cyber infrastructure.
“I think we have to work with the government, but I was wrong about the intensity of the current feeling of mistrust and now I understand something,” he said.
Trust AI control
Basically, Altman and others aligned with him want to work with governments, even as public mistrust grows over the misuse of AI.
“One of the biggest questions the world will have to answer next year is: Are AI companies or governments more powerful? And I think it’s very important for governments to be more powerful,” Altman said. “The future of the world and decisions about the most important elements of national security must be made through a democratically elected process. And the people who have been appointed as part of that process, not me, not the CEO of some other laboratory.”
Altman returned again and again to the question of how the power of AI is coming faster than institutions, governments or most humans can calibrate for it. Systems are becoming more capable and their potential for misuse is growing along with them.
The stakes are getting higher and more serious. The big fights between those who must come up with safe regulations and the companies that, at least in theory, are trying to steer technology in an ethical direction, represent a huge problem.
A diplomatic shrug urging diametrically opposed parties to “find a way to work together” is unlikely to resolve things. Still, it at least means that Altman knows the answer won’t be obvious, even if he phrased it as a request to ChatGPT.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds. Be sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp also.
