On April 29, Senator Bernie Sanders held a public forum on the existential risks of artificial intelligence, calling it “a runaway train with no brakes.” On stage with him were two Chinese AI scientists, Xue Lan (薛澜), a professor at Tsinghua University, and Zeng Yi (曾毅), a researcher at the Beijing Institute of AI Safety and Governance.
Treasury Secretary Scott Bessent, who has been a leading figure in creating Washington’s AI policy, was not impressed. “Instead of harnessing American innovation, Senator Sanders is inviting foreign nationals to tell the United States how to regulate AI. It would be like channeling Hugo Chavez to get advice on how to run our economy — oh wait, the Senator from Vermont did that 20 years ago, too,” Bessent wrote in a post on X.
This week, Bessent’s former advisor and chief speech writer, Sam Lyman, published an investigation arguing that China-linked actors are amplifying data center backlash in the U.S. The report was published by the Bitcoin Policy Institute, a non-partisan think tank that aims to highlight “Bitcoin’s alignment with American interests” and promote policies that “support Bitcoin’s growth.”
The argument that China is influencing American public opinion on AI infrastructure has been picking up steam in conservative circles in the past couple months. “If you’re China, what would you want to do to keep Americans from building data centers, how do you do that? You use our system of free press and freedom of information to influence people,” said Brett Guthrie, chairman of the House Energy and Commerce, speaking in March at the ACA Connects Summit.
The Washington Free Beacon, a conservative media outlet, published an article last month examining coverage from China, Russia and Iran-linked media outlets that emphasize spiking energy prices, local opposition to data centers and the use of old coal-fired plants to power American AI infrastructure.
Lyman’s research pulls from the same set of Chinese state media outputs as the Washington Free Beacon article. “The asymmetry is what gives the operation away,” Lyman noted. “While Beijing’s state media warns American audiences that data centers are environmentally and economically dangerous, the Chinese state subsidizes up to half of the energy costs of its own AI data center operators.”
Lyman also claims that left-wing nonprofits funded by Shanghai-based American millionaire Neville Roy Singham, including the Marxist-oriented think tank Tricontinental and the anti-war organization Code Pink, are parroting Chinese Communist Party talking points on U.S. export controls, the environmental impact of data centers and Big Tech’s contributions to U.S. defense. Lyman filters these narratives through the findings of a 2023 New York Times report that Singham-linked organizations constitute “a global web of Chinese propaganda,” and at various points have attacked Hong Kong pro-democracy activists in London, denied evidence of forced labor in Xinjiang and promoted Chinese lending practices in Africa. Jodie Evans, the co-founder of Code Pink and Singham’s wife, “casts [China] as a defender of the oppressed and a model for economic growth without slavery or war,” wrote the Times.
The Bitcoin Policy Institute report is vulnerable to criticisms beyond its lack of rigorous independent investigation. For one, some experts have questioned whether Chinese state narratives can meaningfully move the needle on this debate. Chinese state media is likely taking advantage of narratives that already exist in the U.S. rather than seeding them, U.S.-China technology scholar Samm Sacks told Semafor earlier this month.
The same Semafor article also highlighted actors shaping public opinion in favor of AI, citing a recent Wired report written by Taylor Lorenz about the nonprofit Build American AI. Linked to tech executives at OpenAI and Palantir, Build American AI is behind a campaign that pays influencers “to spread pro-AI messaging and stoke fears about China,” Lorenz wrote. “[M]any people scrolling their social feeds are likely unaware that they’re absorbing political messaging from corporate interests.”
Critics of AI and U.S. corporate power might view claims of Chinese influence as missing the point. “Telling the hundreds of millions of Americans who are today anti-AI ‘Your opinions were paid for by the CCP’ is not a winning political message,” American Enterprise Institute fellow Ryan Fedasiuk wrote in a post on X, referring to the Chinese Communist Party.
The Wall Street Journal reported on a recent poll conducted by Gregory Ferenstein with researchers at Stanford University and the University of California, Berkeley, which found that “about 30% of Democrats think America should accelerate AI innovation as quickly as possible, compared with roughly half of Republicans and 77% of tech founders.”
If there is a silver lining among these layers of competing information influence, it is that public opinion clearly matters. However it shakes out, this debate will help shape U.S. technology leadership and competition with China for years to come.
Leave a Reply