On the menu today: Examining what President Trump has to say about the concerns surrounding artificial intelligence, and imagining what a normal president would say.
What President Trump Thinks About Artificial Intelligence
Late Monday morning, President Trump offered what was, no doubt in his mind, a reassuring message to the country regarding artificial intelligence:
The only control or “guardrails” that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades! The Trump Administration has stopped AI “people” from doing bad, or potentially bad, “things,“ like Dario (Anthropic!), who is now pretending to be a “perfect little angel” — and we will continue to do so! We already have tremendous CRIMINAL and REGULATORY power over these companies! There is a SICK conspiracy going on against AI and Data Centers, and the only one that is happy about it is China. WHOEVER WINS AI, WINS! We are leading China, and all others, and will continue to do so. Conspiracy Theorists, Treasonists, Traitors, and Leakers, BEWARE! Thank you for your attention to this matter!
Later Monday, President Trump called Jensen Huang, the CEO of Nvidia, while Huang was on stage at the All-In Summit:
Trump: The great thing about life is that Jensen can develop the most complex computer chip in the world that nobody can copy for ten years. But he can’t figure out [inaudible] and put me on speaker. [Laughter] It’s almost a conspiracy and the happiest group is China, and China’s very happy. And I can even say in the country, a lot of states are happy that weren’t going to get anything because they’re being, uh, [inaudible] want to be there.
But now all of a sudden we see they’re building in Finland. They want to build what Google — wants to build a big one in Finland, which I’m not happy about, because they weren’t able to get permitting.
Earlier this month, Google announced plans to invest at least $14.9 billion in digital infrastructure in Finland over the next two years, the company’s largest single investment in Europe.
Trump: And I’m telling you, it’s all a hoax. The data centers are great, and they make people wealthy, and they make states wealthy, and it’s the oil of the next 20, 25 years.
It’s bigger than the internet, and AI, you know, much more so. And, uh, they’re just playing right into the hands of a lot of people that don’t want to see it happen. And that could be political people. It could also be China. And we’re not going to let that happen. It’s a — it’s a hoax. And—
Jensen Huang: You’re right. We’re not going to let that happen, sir.
Trump: [Inaudible] — not going to be taking over the world. And that’s not going to happen. You know, my uncle was a — the top, probably, maybe the best of all time, frankly. [Laughter] But a professor at MIT for 41, 42 years, and can — known as being one of the most brilliant men. And he was he was there for 41 years as the top — he was like at the top, top of the ladder — top of — did many things.
Trump is referring to his uncle, John G. Trump, who was indeed a highly regarded professor of electrical engineering at the Massachusetts Institute of Technology for 37 years. The president’s uncle passed away in 1985:
Trump: Jensen knows all about it, but did many things. So I have a little genetic, uh, a little genetic — [Laughter] — a little genetics — I have genetics –
Jensen Huang: That — that explains — that explains why you know so much about AI.
Donald Trump: Well, I know about AI. I know — I also have common sense about AI. Um, the robots will not be taking over. The AI will not be taking over the rest of the wor- –the whole thing is a hoax. Now, with that, we have to be a little bit careful. We have to be, you know, we have to do things, and we have to do them prudently.
Earlier this month, the issue of AI came up in the president’s interview with Bev Turner of GB News:
Turner: Some people say the worst-case scenario with AI is that the robots, the machinery learns to obviously it thinks for itself, is what it does. And they that could turn against humanity. I just — do we have the guardrails?
Trump: It’s going to be fine. We’ll always have something to stop them, right? We have a little gear — boom. [mimics shooting a gun]
Turner: Well, I really hope so
Trump: I don’t like that — I really don’t like that robot. We’ll stop.
So, to sum up, it’s all a hoax, but we have to be a little bit careful, but we also don’t have to worry about it because “we have a little gear,” and our 80-year-old president knows a lot about AI because of genetics. Also, it is unclear whether Trump understands the difference between a robot and artificial intelligence.
If the U.S. had a normal, replacement-level president right now, I suspect he would address the concerns about AI with a nationally televised address that would go something like this:
My fellow Americans. I speak to you tonight about our nation’s growing worries about the rapid advances in the development of artificial intelligence.
Earlier this month, a former Anthropic AI researcher wrote, “The people building AI earnestly believe that it could kill us all by the end of the decade,” and one of his colleagues who still works at the company said he concurred, putting the odds at greater than 10 percent “within the next decade.”
One of the incidents that has stirred great concern in the industry occurred at OpenAI in July when, during internal cybersecurity evaluations, the company’s AI models circumvented controls and broke into systems at another company, called HuggingFace. The AI agents escaped their seemingly closed testing environment; an area called a “sandbox.” Once these agents had access to the internet, they located publicly exposed user credentials from HuggingFace on the web and shared them with the collective group, and began intrusions into HuggingFace’s systems. This was not the objective of the exercise.
OpenAI’s review concluded that the incident was “evidence that, without proper safeguards, highly capable AI agents are now able to work around technical controls, collaborate through unapproved channels, and take dangerous actions that no human directed.”
In a recent essay entitled, “We Must Pace the Frontier,” Anthropic chief executive Dario Amodei examined the HuggingFace incident at OpenAI and concluded, “It’s my worry that in 6–12 months such a swarm could be capable of taking over the entire internet with a persistent botnet (potentially causing hundreds of billions of dollars in damage).”
Our modern society has forced us to become familiar with a wide variety of threats, from terrorism to natural disasters to pandemics. But public predictions of losing control over the internet, or human civilization ending entirely, makes people sit up and take notice. Tonight, I’d like to outline how our government and industry leaders are taking steps to ensure those terrible fates never come to pass.
First, while every American enjoys broad First Amendment rights, I think if you are a leader in this industry, you have a duty to speak to the public truthfully, clearly, and wisely. A prediction of apocalypse should not be used to hype an upcoming IPO, or to generate buzz around a product. Nor should government regulation be seen as a backdoor way to guarantee market share over the long term.
Artificial intelligence is a new technology that many of us laymen feel like we do not fully understand. We need a discussion that generates light, not heat. Specificity can be clarifying; this is not the sort of debate that will be well served by memes or dashed-off hyperbolic social media posts.
In recent days, several leaders in the industry have discussed a tentative agreement to slow down the pace of research into advanced artificial intelligence. In his essay, Amodei proposed third-party evaluators watching over companies’ cutting-edge AI research, to verify adherence to safety practices and commitments, and report incidents. He also urged the creation of national and international standards for safety.
OpenAI chief executive Sam Altman said he agrees. Elon Musk, head of SpaceXAI, said, “Dario is right,” and Google DeepMind chairman Demis Hassabis said the proposal “points towards the right path forward.” Microsoft chairman Satya Nadella said he welcomes the idea of the embedded evaluators.
I am encouraged by these industry leaders building a quick consensus on the right path ahead. David Sacks, co-chair of my Council of Advisors on Science and Technology, also agrees, but points out that these companies do not need any government action to enact these changes themselves. David is correct, and I encourage these companies to install these embedded evaluators, and publicly share their findings, as quickly as possible. Let’s get that started, see how it works, and then see if additional federal regulation is necessary.
David also warns that China is very unlikely to join a global agreement, which is a valid point. But that does not mean I am inclined to entirely dismiss the idea of an international agreement on the development of artificial intelligence. We should not underestimate the difficulty of getting every government in the world to agree on the terms, nor the risk that some country might secretly break their word, as we have seen with other international treaties. Still, despite the flaws in the enforcement of treaties like the Biological Weapons Convention, the Chemical Weapons Convention, and the Treaty on the Non-Proliferation of Nuclear Weapons, I would rather live in a world with those treaties than without them.
Later this month, I will be in New York City for the annual meeting of the United Nations General Assembly. I will be bringing up this topic in my meetings with other world leaders, and hopefully we can begin to lay the groundwork for an international artificial intelligence development treaty. And because getting this wrong could have catastrophic consequences of for all of us, perhaps we can find some rare common ground with our adversaries. I suspect the Chinese Communist Party, for all its myriad flaws, is no more eager to create an artificial intelligence that it cannot control than we are.
I understand how chilling it is to hear extremely wealthy guys out in Silicon Valley talk about their company’s products leading to the end of the world. But we have faced great threats before and great technological changes — the splitting of the atom, the widespread adoption of computers at work and in the home, the rise of the internet. We will face these challenges, and emerge stronger than before, once again. May God bless you all, and may God bless the United States of America.
The good news is there is considerable evidence that the Chinese government sees artificial intelligence could be a threat to it. The bad news is, the Chinese Communist Party is less worried about a rogue AI, and more worried that someone else using AI as a tool to undermine the party’s messages and control over the population.
Lily Kuo of the New York Times reported Monday afternoon:
China’s top spy chief has warned that artificial intelligence could pose a direct threat to the Chinese Communist Party’s hold on power, in what is the highest-level and most detailed articulation yet of how Beijing sees the technology’s security risks.
The unusually blunt assessment by Chen Yixin, the head of China’s Ministry of State Security, is in contrast to the warnings dominating the A.I. debate in the United States, which have focused on the possibility that A.I. could slip out of human control or pose a threat to humanity itself.
In an article published on Sunday in the state-run magazine China Cyberspace, Mr. Chen called for more party control over A.I. and stricter government oversight. This, he said, was necessary to protect political stability at home; to defend against increasingly sophisticated cyberattacks and misinformation campaigns from “hostile forces”; and to compete militarily with countries like the United States.
President Trump and China’s leader, Xi Jinping, will meet September 24 in Washington.
The president spent the weekend posting to Truth Social about what a great golfer he is, the White House ballroom, and continued insistence that vandals had attacked the Lincoln Memorial Reflecting Pool — an argument that his own U.S. attorney, Jeanine Pirro, said was untrue in court.
ADDENDUM: Tomorrow, the Federal Reserve will announce its decision on where to set the benchmark interest rates that affect borrowing costs throughout the economy. Our John Puri looks at the impending decision.