And so just given all these threats and given your responsibility to infuse the Department of War with this type of technology, how can you do that given those kind of concerns, especially if models may already be compromised?

Well, I was saying that that's what the model companies themselves say about their products.

And I think it is certainly true that there are cyber capabilities that have some risks if someone's going to use them offensively.

I think that we can sort of all agree on.

And then we're also then, therefore, in the midst of a cyber hardening exercise.

We're not only Department of War, but government and businesses are going back on their old code and their old systems and making sure there's no vulnerabilities and so on.

But that's what they say about them.

And then just this question around, you know, are the existing laws sufficient for us to deal with the consequences of, I don't know, runaway AI or AI that's doing something it's not supposed to do?

Emil, what do you want folks to understand about the superintelligence force? Is it about sort of looking ahead to future risks or investigating things we already know something about, like Hugging Face?

It's more about looking toward future risk and future opportunity.

And the thing is that there was a lot of scare tactics that have been perpetrated in the last few weeks, potentially with some foreign influence, that sort of scared the American people.

You see how this polls.

And the idea is to make sure that American people understand that AI can work for them, can support their goals, and to make sure that we're talking to the different groups of people who it might benefit.

but also to be cognizant of the risks and make sure that the AI companies, the mega corporations are taking their own stock of the risks and investing and protecting against those risks and ensuring that we have the right regulations, laws and powers that we already have in place and make sure they understand that to ensure that they know that they're liable in certain cases.

And they take those risks seriously so that we have a robust AI industry in the United States that beats China, where the risks are being borne by those who create those risks and ensuring that the interests of the American people are first and foremost.

But, I mean, is any liability being enforced?

What was your reaction to that? How should we view that moment in time, given it does seem to have sparked a lot of this current conversation?

Yeah, it's the right question.

I read that extensively.

I would say that it was in their testing environment.

Their testing environment, which is the same one the Anthropic uses from a company called Irregular, which has done a lot of these testing environments, clearly is not doing a good job because they're supposed to constrain what they're testing.

Now, the escape by that agent that attacked Hugging Face is concerning.

And it just shows to me they're not investing enough in that side of the equation.

How much should we be pushing back on China to try to stop their progress?

Interesting.

What do EAs wanna do? They wanna control the w- the system of government we have.

They want to give it to a machine god.

[laughs] So I w- I got the AI portfolio six months into the job.

I opened, I, "Let me read the contracts here.

I'm a, a former lawyer," or I went to law school.

I didn't practice law.

Oh.

Why, why don't you explain how you're kind of getting at that supply chain issue and how we make sure we can manufacture these drones here in America at, at the rate we need?

Yeah, that's, that's right.

I would say we haven't been at, at, at it this long.

Only this administration took it seriously in the last year and a half.

And what we've done is, uh, we've tried to stimulate and begun to stimulate the domestic production of the components and the all-up systems for these drone manufacturers in the last year and a half.

And we've done things like loaned $1.800 million to Performance Drone Works, who's gonna make components that are otherwise, um, only available from China or from other adversaries.

And we're doing that across the domestic supply chain to make sure that they're building drones here.

What he, his point is that it's, it's important for venture capitalists to be able to say, "Here are standards that have been set by the government, and if we see whether a company can perform to those standards or not, it gives us a better idea of where to invest."

I mean, tell me the department side of this story.

You know, it's sort of odd in that from a domestic surveillance.

The Department of Work doesn't do domestic surveillance.

We're not a law enforcement agency.

We have intelligence agencies that look at our adversaries, but it's not in our mission set.

In fact, there are incredible procedures to ensure that we don't touch Americans.

There's the National Security Act of 1947.

How do you think about that? I mean, I think

Mm-hmm

...

ChatGPT and the subscriptions.

Um, xAI sort of hadn't been s- started really until 18, 24 months ago, and then Google was also focused more on the consumer.

They were focused more on enterprise.

When I say enterprise, I mean enterprise at large, an enterprise like the Department of War or an enterprise like a big company.

Um, and so they were naturally started sooner here.

So talk a little bit about how you're working to reform the procurement process and why, why that's gonna be good for people.

Can you, like, maybe draw, like, what are the things that you really care about in your job today?

I care about a few things.

I'll say that one of the things that I've been spending the most time of late on is trying to infuse AI into the Department of War.

The Department of War is 3 million people.

It's literally the biggest organization in the world.

And therefore it has all the opportunities and challenges you find with how do you make a 3 million person organization that has technology that's 20 years old from a sort of productivity standpoint, how do you make it more efficient? And then actually people are happier if you could, you know, take the mundane tasks off their hands.

So there's the enterprise level of how you can use AI here, just like any big organization would.

How far off am I?

How do you prioritize? How do you take stock of where those priorities are? How do you figure out? I mean, if you read the news, it feels like there's an endless list of things that we need to catch up on.

I-- When I took the role in May, I was sworn in.But did the inventory like any new leader would do, and we had fourteen critical priority areas in this role, and my predecessors had ten, and then someone added four, so there were fourteen, and they hadn't changed in nearly a decade.

So I get to look at them and say, "Well, okay, well, why are these critical? And by the way, who can remember fourteen things when you're trying to motivate a workforce? And what is an integrated network systems of systems?" They were sort of techno babble.

So I said, "Well, let's study and say what's really important." And we got them down to six, and they were the places where I thought we had the greatest opportunity for change and for growth and impact on our combat power and for our industrial base.

And that was the starting point.

And applied AI was number one.

So then we moved the chief digital and AI office into my group, and we were able to run extremely fast with that 'cause we're way behind in AI at the department, and we know it, it's penetrating the rest of the world so dramatically, so quickly, and our adversaries were also using it for lots of different purposes because they have less trust in their command and control, so they wanna use computers to sort of eliminate human decision-making.

We want the best capabilities." What do, what do you need from the companies and the startups?

Is, is that the case?

No.

Uh, it's sort of a red herring.

When we got the contract that we inherited from the Biden administration, there were 25 pages of different restrictions on the Department of War and its use cases.

Uh, and they were so restrictive when I looked at them that we started negotiations to sort of reset the baseline.

I mean, we are the Department of War.

We use their, uh, software and services and hardware for Department of War activities, and we wanted to baseline where we're at.

What, what do you say to, to that argument?

You can't do a war game scenario with it?

You could do a scenario, but you can't...

Like, let's suppose you're writing a plan saying, like, "If this happens, here's what we would do," and it might involve a kinetic strike which causes harm to a human.

So you're like, "Well, what do you think these folks do?" You know, this is the Department of War.

This is what we do.

And so I said, "Okay, well I, I've gotta, number one, have direct relationships with these companies, not just through Palantir, 'cause I wanna use it more broadly.

And then number two, I need to have the terms of service be rational relative to our mission set." So we started these negotiations and, and took three months, and I had to sort of give them scenarios about, like, this Chinese hypersonic missile example.

Where are we with respect to where we need to get to in the US manufacturing supply chain?

What is it? And there's obviously some new technology here in the case of what happened in Venezuela.

Besides the discombobulator, what we've got is a very well-trained military.

The global war on terror was a disaster in so many respects, but- The people now who are fighting that are generals now.

And so they've learned a lot of lessons.

And you compare that to the Chinese military.

They don't have a lot of experience.

In fact, the decapitation they did in the Chinese military, the one guy they took out was the one guy who had experience in Vietnam.

a war game scenario with it?

Your response?

Yeah.

Well, so we started...

When I got the mandate to be the chief AI officer for the department, essentially, um, I was looking across where we had deployed AI, and it was very piecemeal in very different parts of the organization, and I wanted to make a big push to make it enterprise-wide.

So we restarted talks with all the companies, and we launched Google Gemini on our unclassified networks.

And out of three million people, over a million two have used it so far in just two months.

And then we signed a deal with xAI across both unclass and classified levels.

Do we have enough, and how do we counter things like drones?

What were the concessions?

Yeah, the concessions are pr-pretty simple.

Uh, we agreed in writing to, uh, ensure that the Department of War, War was following all laws and regulations, including the National Security Act of nineteen forty-seven, the FISA Act, and all other applicable laws and regulations, because mass surveillance of Americans is already illegal.

So we were, uh, offering to put all of that language and affirm that we were following all laws in the contract.

When it comes to autonomous weapons, similarly, we said we'll follow all laws, including a DoD directive that's been in place for years that governs how we would use any such weapons.

And then we affirmed that there would be human oversight, uh, over all kind-- of every kind of, uh, part of the development process or engagement process or use of autonomous weapons.

And he didn't like the word, I guess, as appropriate at the end of that sentence.

Under Secretary Michael, uh, is lethal autonomy really so critical for, uh, future national security?

What were the concessions?

Yeah, the concessions are pretty simple.

We agreed in writing to ensure that the Department of War was following all laws and regulations, including the National Security Act of 1947, the FISA Act, and all other applicable laws and regulations, because mass surveillance of Americans is already illegal.

So we were... offering to put all of that language and affirm that we were following all laws in the contract.

When it comes to autonomous weapons, similarly, we said we'll follow all laws, including a DOD directive that's been in place for years that governs how we would use any such weapons.

And then we affirmed that there would be human oversight over every kind of part of the development process or engagement process or use of autonomous weapons.

And he didn't like the word, I guess, as appropriate at the end of that sentence.

Under Secretary Michael, is lethal autonomy really so critical for future national security?

Mm-hmm.

So I, [chuckles] I looked at them, I said: Well, the, uh, number one, that if, if there's that many, they can't be critical.

But number two, hard for people to hold in their head that many priorities and wake up every morning knowing that they have to get up and execute against those priorities and make progress.

'Cause you get distracted, you get too much, uh, disper, you know, uh, separation, and therefore not enough resource to any one thing.

So I, I cut 'em down to six, a- after studying them, and I made it more action-oriented, so we put sprints behind them, like you would in an engineering team.

So, uh, you know, off the top of my head, the, the number one is applied AI, because we're not building a foundation model at the DOW, because there's being hundreds of billions of dollars being spent by the private sector on this.

So how do I adapt or use what's being developed in the, in the private sector and apply it to the Department of War use cases? So that's number one.

Could you talk about, more broadly, what are the set of applied AI initiatives that you all are focused on? I know it's a major thrust, and it's one of the key areas that you talked about, so.

Why did you choose these six and where are you seeing the most impact?

So this, the two predecessors ago for the job I have now, they added 10.

And that was in 2018.

And then the next person added four more, so there were 14.

So these critical technology areas were almost 10 years old in 2018.

And if something's critical for 10 years, that means you're not making much progress on it.

So that's number one.

How is that being implemented and is this going well? Is this like picking up steam? Like how have you seen initiative versus action and like the gap between the two?

Mm-hmm.

And, uh, like, uh, the pitches, uh, you know, okay, you have to have some patriotic instinct obviously, but then maybe you're in between companies or you're just motivated.

I mean, Elon has five companies and he still works for the government for, like, seven months and brought a ton of Doge people who were motivated by the mission.

So I'm trying to motivate people by, "Hey, this is the biggest technology deployment in the world.

There is no bigger organization than the Department of War.

There's no more exquisite, crazy interesting use cases whether they're intelligence, war fighting, you know, even corporate use cases and you'll get to be a part of creating that.

It's never been done before." That's a pretty good pitch.

Uh, do you think a lot of other kind of groups need to kinda pursue a project like this? And are you guys trying to help create kind of a framework so others can, can benefit?