Why DC Views Superintelligence as a National Security Threat
MTS
October 15, 2026
Sam Hammond, Director of AI Policy at the Foundation for American Innovation, discusses Washington's shift toward treating AI as a national security issue through the Superintelligence Task Force, model weight security, and antitrust defenses for lab coordination.
Sam Hammond
It's no longer, setting aside David Sacks, it's the archetype of treating AI policy as a thing for venture capitalists, and less of a national security issue. This is now squarely within the realm of the national security state and the Defense Department.
SPEAKER_2
Hello, everyone, and welcome back to MTS. Today, we are joined here in person, live with Sam Hammond, who is the Director of AI Policy and Chief Economist at the Foundation for American Innovation. Really excited to talk to you. I know you do a lot of work in DC, particularly about thinking about how governments can actually be informed when it comes to making policy around Frontier AI. And so, wanted to get your takes on everything that's happening in Washington, the new initiatives, the new policies, just what needs to be fixed, how Frontier Labs can actually coordinate and work together and govern, all of this fun stuff. So, Sam, welcome to MTS.
Sam Hammond
Thank you for having me. Yeah, and we're actually debating whether I should be the Director of SI Policy now.
SPEAKER_2
Really, I know, we're not even allowed to say AI anymore, and so apparently, we're the enemy.
Sam Hammond
Yeah.
SPEAKER_2
What do you think about that? This actual take on changing artificial intelligence to superintelligence?
Sam Hammond
I think it's going to be accurate soon enough, right? And I think this is one of the bigger... So, the positive version of this is the things I work on really are like superintelligence policy, frontier AI policy, is not worrying about Netflix recommendation algorithms.
SPEAKER_2
Yeah.
Sam Hammond
The negative side of this is that the way that they're going about this name change through executive order has to be coextensive with the existing definition within the National AI Initiative Act, which is one of the first big AI laws from several years ago, and that defines AI basically as algorithms. So you get this funny situation where the other day, the Southern District of New York posted a headline that was like, North Carolina man arrested for superintelligence assisted music streaming fraud.
This person, whatever was going on, they were probably cloning Taylor Swift songs and stealing streaming revenue, but that's not artificial superintelligence. So it's just funny, definitely in this no more right now, but I think it would be appropriate to call me the SI Policy Director because I don't work on not superintelligence.
SPEAKER_2
Yeah. It's pretty interesting. You're focusing on superintelligence, you're focusing on frontier and AIs impact on the frontier, not necessarily algorithms, not necessarily some of the smaller things that are happening. I am curious, though, just generally in DC, whether our policymakers have enough knowledge and context to really be making effective decisions about policy and governing superintelligence.
Yeah.
Sam Hammond
I would say everything's relative. I think it's been said before, but DC is like the second most AGI-pilled city in America, probably on Earth.
SPEAKER_2
After SF?
Sam Hammond
After SF, obviously.
SPEAKER_2
Of course.
Sam Hammond
And I think we've had some role in that, like FAI, we were founded initially as a networking events organization to try to bridge the cultures of Silicon Valley and DC back in 2014, sort of like Hill and Valley before Hill and Valley.
And I run a conservative AI policy fellowship where essentially I just tried to expose Republican Hill staffers, administrative officials to all the core issues in AI policy, including like trying to understand forecasts, the trajectories. You know, back in the day, like the famous meter graph, which is now obsolete. And so like DC is very aware, probably not all the way aware, but the hugging face, Superintelligence Task Force, these things are representing real sort of more than a vibe shift, but like actual sort of updating on the part of lawmakers and policy makers. And partly that's informed by like the quality of their staff. So, you know, we do a lot of talent development work. There's places like Horizon, AAAS, Tech Congress, where we've served places that have sort of cultivated networks of young and technical talent to go work in Congress, go work in the administration first and beef up the situational awareness. There's still a lot to do, obviously.
SPEAKER_2
There's still a lot to do. And I want to talk a little bit more also on that talent piece in a second. Before we get into that, what you talked about earlier around the Superintelligence Task Force, this is something famously earlier this month that was talked about and how it actually would play out.
How does this work? I know there's this 120 day window. What actually happens within those 120 days? What's the overall goal of the Superintelligence Task Force? What's happening there?
Read the other 20 minutes of this episode here.
$15 gets you this episode and 99 more. Credits never expire. After checkout, one link brings you back here to read it.
Not what you expected? Full refund within 14 days.
Thousands of transcripts fetched by people building searchable podcast archives
Fetch transcripts with code
The same credits work from a script or an AI agent. The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000796438267