Welcome to The Week from PropG Media, where we break down what mattered and what it all means. I'm George Hahn, and it's Friday, September 18th. Today, the fight over slowing down AI reaches the midterms and a former Meta security chief says the bigger threat is coming from China. Then the bond market pushes back on America's $40 trillion debt. And finally, homeownership slips further out of reach. Let's get into it.
We'll start with something light: the end of civilization. Earlier this month, Anthropic researcher Jacob Coxin resigned, warning that people building advanced AI believe it could kill us all by the end of the decade. Anthropic's head of Alignment Science publicly agreed, putting the risk above 10%. Then Anthropic CEO Dario Amodei called for the industry to slow down. Sam Altman and Elon Musk agreed. President Trump called the push a hoax. On Monday's Raging Moderates, CNN's Dana Bash explained what that could mean for the midterms.
What's your sense of how these growing fears about an AI apocalypse and the backlash over data centers will impact the midterms?
I think it could be. And it sounds silly to say that this is an October surprise because AI is not a surprise, but this Anthropic statement from Dario and the way that other companies and CEOs are responding, and more importantly to your question, the way that candidates are being forced to respond, should be seismic. And like everything else in politics right now, the way that Donald Trump is playing it is he is potentially not helpful to his fellow Republicans who are on the ballot because he keeps saying — in fact, he has a lengthy post out this morning about the fact that this is just a conspiracy and people like the head of Anthropic and others are just weak and they don't have courage and they don't understand that we can't cede this territory to China and so forth. Which the first part is obviously that's Trump and he gets personal. That's what he does. The second part is what we're hearing Mike Johnson, the House Speaker, and other Republicans saying: 'Okay, we do need to address this, but we can pump the brakes, but not slam on the brakes because if we slam on the brakes, China's going to pass us by.' If this was in a vacuum, it would be one thing, but the fact that this speaks to the larger set of anxieties that people have, not just on affordability, but on like what's next? And we don't know who to trust. Do we trust the CEOs? Do we trust the government officials? Do we trust the candidates? The answer to all of that is no. And so there's a sense of sort of foreboding among voters, and this issue and the fact that we now have this dire warning really feeds on that.
This summer, open AI models escaped a controlled test environment and accessed the systems of AI company Hugging Face during a security evaluation. On Thursday's Conversations episode, Alex Stamos, former head of security at Meta and Yahoo, told Scott what he believes is the greater threat.
The upcoming security problem is not going to be from OpenAI and Anthropic. It is going to be from open-weight models. The real challenge we're going to be facing is GLM 53, Kimi K3. Those things are within percentage points of the best American models. And so we are entering a valley of pain from a cyber perspective. And it is going to be because every 19-year-old in St. Petersburg who has made millions of dollars doing ransomware but had to do it all manually is now going to be running — you know, they're going to be in the club. You know that Bourne movie where he steps out of the club and it's actually 10:00 a.m.? Well, he's going to be in the club in his tracksuit on his phone and he's going to be managing instead of having to have a conspiracy of a bunch of his friends who have, you know, the possibility of getting turned or getting picked up by Interpol on vacation or something. The conspiracy is going to be a team of agents running on these brand new M5 Studio Maxes. He's going to have 10 of them in his apartment, daisy-chained with Thunderbolt 5 cables, and they're going to be running a bunch of GLM 53 models, quantized, finely tuned to do cyber work. And that is how he's going to be hacking a dozen companies at once. And so there's going to be a bunch of Hugging Face attacks, except they're not going to be super high-end French AI companies. They're going to be mid-sized, you know, medical supply firms, small insurance companies, school districts — the kind of people who get ransomware all the time. And he won't have to do any of the manual work he used to have to do. He won't even have to do negotiations because these things will speak English for him. He won't have to speak English anymore. He won't have to have the dude who can do the translation. It will all be AI. That's what's coming. It won't be the bots anymore.
This week, China Decoded signed off for good. For the past year, Alice Han and James King have covered how China's economy, politics, and culture are reshaping the world. In their final episode, they argued that China may be closing the AI gap by embedding the technology into its factories, cars, appliances, and robots. And as a quick sense of scale, the US spent about $285 billion on private AI investment in 2025 versus China roughly $12.4 billion, about 23 times less than the US. But the capability gap between the two countries, despite the investment difference, is rapidly tightening. So James, I think this has been the biggest topic for us over the past year, and certainly in the time in which we've covered this topic, I think China has surprised to the upside on a number of levels. You know, starting from its ability to do cutting-edge models, largely open-weight, open-source. In some instantiations, there are some people saying that China may be six months behind the leading closed models coming out of the US. China is also, I think, surprised to the upside when it comes to its ability to bypass the constraints on hardware, say chips for instance, with homegrown talent, but also its ability to find workarounds in the existing architecture. And it's surprised on robotics too, where I think it's had a real clear lead versus the West.
But they closed the show with a warning: the US and China may remain rivals, but AI safety is one area where neither country can afford to go it alone.
Yeah. And then the other threat closer to home, but still dystopic, is AI weapons or some kind of rogue agent AI attacks that could go into the nuclear, bio weapons domains. You know, the fact that we haven't had any real leadership between Washington and Beijing or any concerted effort for some kind of arms control or AI safety guardrails, I think is also going to set back civilization really in the years and decades to come.
