Former President Donald Trump's position on the prominent AI startup Anthropic appears to have softened recently, yet this does not signal a broader relaxation of Washington's vigilance towards advanced artificial intelligence.
In a recent media interview, when asked if he considered Anthropic and its CEO Dario Amodei a threat to U.S. national security, Trump stated that he does not see them as a threat now, but admitted he might have thought so a week ago. He noted that the company responded swiftly after the U.S. government expressed concerns, acting in a "very responsible" manner. While he stated he would not shut the company down, he added that he was not certain whether he would be compelled to use emergency legal powers to intervene.
This shift suggests that while the administration's tone has temporarily moderated, it has not ruled out the possibility of future direct intervention concerning Anthropic or the broader AI sector.
Observers suggest that the rapid change in attitude towards Anthropic over just one week indicates the administration is re-evaluating the AI industry through a national security lens, attempting to balance maintaining America's competitive edge with mitigating risks associated with technology proliferation.
During the interview, Trump revealed he met with Amodei and other tech executives at the G7 summit in France. He acknowledged that Anthropic responded quickly to government worries. "He responded to us very quickly because it involves tremendous responsibility," Trump said.
However, Trump did not fully dispel his concerns. He clarified that he would not shut down Anthropic, but added, "I would do that, but I'm not sure I have to do that. I think so far, it's been very responsible." This indicates that while no immediate further action is planned, the option for necessary intervention remains on the table.
Furthermore, Trump did not exclude the possibility of utilizing emergency powers granted by the Defense Production Act (DPA) to compel Anthropic's cooperation with the government.
This stance comes just days after the U.S. Commerce Department mandated that Anthropic must obtain prior U.S. government approval before allowing foreign individuals, companies, or governments access to its most advanced Fable 5 and Mythos 5 models. This move is seen as the most significant intervention by the U.S. government into an AI company's operations to date.
Reports have indicated that within the administration, there is concern that the most advanced AI models could become strategic assets akin to nuclear technology and advanced semiconductors, necessitating restrictions on their overseas dissemination.
In the Friday interview, Trump remarked, "We didn't like some of the things they were doing before, but they responded very quickly." He specifically praised Amodei as "very smart" and said Anthropic has been performing well so far.
Trump emphasized that the United States is leading other nations in the AI race and therefore would not lightly restrict the development of domestic companies. However, this does not mean the government will forgo regulation. Trump believes the AI field involves "tremendous responsibility," and the government must ensure the technology remains under control.
Speaking about the DPA, Trump said, "I have a lot of tools at my disposal," but added, "I'm not sure I have to use it." Addressing concerns raised by Amazon, Trump noted that Amazon, "as both a competitor and a partial shareholder, reported on Anthropic. They were unhappy with what Anthropic was doing and were very concerned. I think so far, everyone has handled it very responsibly."
Analysts believe the change in the administration's attitude towards Anthropic does not signify a relaxation of oversight. Instead, it sends a clearer signal: as long as companies are willing to cooperate with government demands, the U.S. will support their development. However, if the administration determines that cutting-edge models could threaten national security, it will not hesitate to use executive power to intervene.
For Anthropic, the timing of this moderated stance is opportune. Reports indicate that Anthropic confidentially filed for an IPO several weeks ago, with the company's latest valuation exceeding $900 billion. Previously, the Commerce Department's restrictions had sparked market concerns that if the White House continued to view Anthropic as a national security risk, the company's commercial prospects and listing plans could be impacted.
Now, Trump's public statement that he does "not consider Anthropic a threat" alleviates some external concerns in the short term. However, his assertion that he would "take action if necessary" means the sword of regulation still hangs over the AI industry.
As the U.S. government increasingly views the most advanced AI models as national strategic assets, future AI giants like OpenAI, Anthropic, and Google may face a regulatory framework similar to that of the semiconductor industry. For the administration, while ensuring America wins the AI race is crucial, preventing technology proliferation and controlling potential risks is becoming an equally prioritized strategic objective.
Comments