Today on the AI Daily Brief, we are discussing a question that is extremely easy to ask and much more difficult to answer.

Who controls AI?

The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI.

All right, friends, quick announcements before we dive in.

First of all, thank you to today's sponsors, KPMG, Insightwise, AIUC, and Blitzy. To get an ad-free version of the show, go to patreon.com slash AI Daily Brief, or you can subscribe on Apple Podcasts. To learn more about sponsoring the show, send us a note at sponsors at AIdailydlybrief.aI. While you're on AIDilyilybrief.ai, you can also find out about the other projects in the AIDB ecosystem, including Claw, Enterprise Claw, registration for which is going on right now, basically if your enterprise wants to learn how to build agents and agent teams, or just more podcast-related stuff like subscribing to the newsletter, which is newly rebooted.

Now, if you've been listening to this week, you'll know that we had something of a time of it getting back from South America. Door-to-door it ended up being about 55 hours, and that didn't include the seven hours that it took me to go drop off the rental car and pick up our old car, which was sitting at the airport parking lot. In any case, because of that, I had to miss Wednesday's show, not something that I do very lightly, and so as a makeup, I had slated to do an extra show over the weekend on the day that I'm usually off. As it turns out, this was a pretty opportune week to have that slot open, because my goodness, as Ron Burgundy would say, boy, that escalated quickly. I'm referring, of course, to the skirmish-turned all-out war between Anthropic and the Pentagon that came to a crescendo in a head on Friday night.

The TLDR of what happened is that not only did the Trump administration decide to decline to

work with Anthropic, they are attacking them in ways that go far beyond just declining to do

business with them. Now, for the necessary background and to get caught up with the story from where we left it, we actually have to go back to Thursday, when Anthropic CEO Dario Amade released a statement about the dispute. Earlier in the week, you'll remember, Defense Secretary Pete Hegeseth had given Amade an ultimatum, removed terms of use limits by Friday or be blacklisted

from the entire military supply chain. Anthropics red lines were that Claude should not be used

for domestic surveillance of Americans or for powering autonomous weapons. Their stated view was

that Claude is not reliable enough to power autonomous weaponry, and that AI surveillance is

undemocratic and, perhaps more pertinently, has underdeveloped legal safeguards.

The White House's position, meanwhile, was that a technology company should not be dictating

how the U.S. government uses that technology and should be fine accepting terminology that allows

the U.S. government to use it for all legal uses.

Dario's post from Thursday begins, I believe deeply in the existential importance of using AI to defend

the United States and other democracies and to defeat our autocratic adversaries. And it is worth

noting here, especially if and as this conversation gets caught up in broader partisan talking

points, historically speaking, Anthropic has been more vocal about things like China not having

access to advanced technology

than some of their peers. Whereas some of the other AI companies have been either fine with

...