Raj Shah, the co-founder and CEO of Arceo Analytics — a new artificial intelligence company — discusses his new firm, defense innovation and his tenure at the Pentagon’s Defense Innovation Unit Experimental with Defense & Aerospace Report Editor Vago Muradian at the Reagan National Defense Forum at the Ronald Reagan Presidential Library in Simi Valley, Calif. Our coverage was sponsored by L3 Technologies and Leonardo DRS.
Raj Shah
CEO of Arceo
Reagan National Defense Forum
November 2018
Vago Muradian: Welcome to the Defense and Aerospace Report. I’m Vago Muradian here at the Reagan National Defense Forum at the Ronald Reagan Presidential Library in Simi Valley, California, the nation’s leading gathering of military, civilian, economic as well as industry leaders, as well as thought leaders from around the world. Our coverage here is sponsored by Leonardo DRS and L3 Technologies. And we’re talking to Raj Shah, the former head of the Defense Innovation Unit Experimental. The X has since been dropped. You left the job in March, and now you’ve signed up as the CEO for Arceo which is an innovative artificial intelligence company.
Raj, it’s a pleasure seeing you, and I love your tie.
Mr. Raj Shah: It’s good seeing you again, Vago. As always.
Mr. Muradian: So let’s talk a little bit about your new enterprise. Talk to us a little bit about what Arceo is and how you guys are going to change the world.
Mr. Shah: Sure. Arceo.AI is using artificial intelligence to help solve cybersecurity holistically for an enterprise. Right now if you’re a CEO of a company you don’t know how much you should be spending, you don’t know your ROI, and you’re stuck with all your cyber risk, and there’s too many vendors out there.
So what we’re trying to create is basically the HMO of cybersecurity. Where you go to a certain enterprise, they’ll manage your health, and if anything goes wrong they’ll take care of all the financial obligations.
Mr. Muradian: So let’s talk about the organismic solution that’s required for this. Right? I mean one of the things that comes through in so many conversations is, you know, persistent Russian and Chinese penetrations of systems. The most classified systems we may have been breached. Electronic warfare files, which is essentially the library of everything we know about our adversaries have been compromised.
Talk to us about how there needs to be a somewhat much more organismic solution — everybody’s been talking about this, but at the end of the day those penetrations are happening and there are even more sort of organic software vulnerabilities that people are exploiting. What are some of the right ways to think about this? As self-serving as your answer might be ultimately, on how to sort of fix this problem.
Mr. Shah: I would step back and first say there’s two categories of threats here, right? So one, you’ve got high-end nation-state type of attacks. That is where the U.S. government and our intelligence community are the best at what they do. But if you look at the type of attacks that are in the news, that are affecting our types of records, credit cards, even the defense industrial base —
Mr. Muradian: Or the Marriot hack as we just saw.
Mr. Shah: Exactly. If you look at those, most of those attacks are not because of some nation-state news piece of zero-day, but simple blocking and tackling. It’s not did you have a firewall. Did you plug it in? Did you configure it right?
I think hackers are also lazy, and they’re going to find the easiest way to enter the door to the house they want to get in. And right now we’re in a situation where people are trying to put more locks on the front door, but all the windows in the back yard are open. How do we monitor that? How do we do it at scale? And that’s where technologies such as artificial intelligence can play a big role.
Mr. Muradian: There’s a lot of debate about sort of the state of AI, where the United States falls in this sort of arms race. The Russians, Vladimir Putin has made a statement it’s important for the nation to be a leader in this from a national security perspective. China has set a national policy, right, and Xi is now a paramount leader, and so he’s said look, 20 percent of the nation’s R&D is going to go to this capability which has a tendency of getting people’s attention.
How do you rate relatively where the United States falls in the artificial intelligence universe, given that that is really increasingly being — that, you know, big data are being, and quantum computing and quantum communications, in fact, are sort of being seen as sort of the handful of holy grails for the future.
Mr. Shah: Yeah. I think there’s a big debate between are open societies or closed societies going to be victorious in AI and who’s going to get the advantage?
Right now in AI, there is the question, is it the algorithm that is the most important thing? Or is it data that’s most important? Unfortunately closed societies, authoritarian societies, have access to more data because privacy doesn’t exist. We, as the Western world, have free and open academic institutions, and that’s why we have better algorithms.
I think it remains to be seen which one’s going to prevail, though. A lot of folks think it’s about the data side. We can use our own advantages as a Western world. We like to attract talent here. We have an open debate on things. That’s what’s going to be victorious for AI.
Now if you think about those countries you named in particular. We’re here at the Reagan Forum. I think AI and Xi learning has been in every panel innovation has been talked about. Right? And it’s wonderful that the awareness is being raised. But now it’s time for action.
For example, the Department of Defense has something called a Joint AI Center. I think it’s a wonderful thing. They have DIUx or DIU now. A wonderful thing. But let’s look at the budgets. The Joint AI Center, which is the premier AI effort in the department, has about $70 million in the last budget cycle. If you look at China in 2025, it’s measured in the tens of billions. We’re multiple orders of magnitude away so I think we have the right vector, we need significantly more thrust and have our own Sputnik moment here in the U.S.
Mr. Muradian: And for anybody who doesn’t know, Raj is a very distinguished fighter pilot, at the end of the day an F-16 driver with many humorous stories, but we digress.
One of the things about fighter pilots is how very hard you can self-grade your own performance.
Mr. Shah: Yes.
Mr. Muradian: You were the second person to occupy the job at DIUx. Very forward, very engaging in terms of always trying to tell the story and the message to advance the organization and to grow it.
As you look back on your tenure, how would you grade the progress you and the department made in inculcating these lessons and actually moving the innovation needle, despite all of the talk about it. You know, there is a lot of debate about how much the needle ultimately ended up moving.
Mr. Shah: Yeah, a lot of challenges in standing up DIUx as it was. It was trying to get the Pentagon to understand how innovative technologies can help our men and women on the battlefield. It’s helping Congress understand how they can have adequate oversight and yet move quickly. And of course, the messaging in places like the Valley of why companies want to work with the department.
I think if I were to step back and say how did we achieve across all those three? If I look at the department, they’ve dropped the X. They’ve realized that these types of technologies are here to stay. It’s wonderful.
If I look at Congress, we went from a zero [IS] budget to increasing budgets.
Then in the Valley, you’re starting to see companies and investors look to build companies that are focused on the Defense Department. We, in the course of the time there, did about 200 million and I think it was about 65 contracts that were given out. Which is great and it’s a wonderful start. But again, if I think about the scale of the problem, it still has a ways to go.
The good news and the really great news is they’ve got a wonderful and amazing new leader in Mike Brown. He has the right vision and the right energy to continue to grow it, so I’m actually quite optimistic that the foundations that me and my team were able to build up will get to scale and grow.
Mr. Muradian: And of course Mike Brown, former CEO of Symantec, coming with just an extraordinary amount of experience that he’s bringing to the job.
Mr. Shah: That’s great. It’s wonderful to have someone that really understands the value and has been a trusted and demonstrated leader there, now back in public service. He is just one example of a slew of talented people that want to come and help the Defense Department and they believe in the mission.
Mr. Muradian: So let me ask you this, right? You were somebody who had military experience, a fighter pilot, but you were very much of Silicon Valley. And yet there’s a lot of debate about whether or not these transnational global companies are going to serve the national interest even though they may be American companies. Their shareholding and their business bases are global. And so some folks have observed that okay, so Google, China okay; but DoD not good. Obviously talking about the Maven controversy.
First, how big of a challenge and problem is this? And second, you know, how would you sort of, what are better ways that the department can engage with some of these transnational companies at the end of the day, given an understanding of what’s driving them on the one hand, whereas what you want out of them from a DoD perspective?
Mr. Shah: Vago, I love speaking with you. Right? You’re I think one of the good guys in the press. Unfortunately, most of the press is very focused on controversy and what is going to get attention. Yes, there were some serious, people with some serious concerns at places like Google with what we’re doing. But I don’t think that that’s the shared view across the Valley.
In my experience, both when I was at DIUx and now, is that most people in these places are patriots. They do want to help. But they want to work in a transparent fashion with the government. Right? We’ve seen different actions from companies like Microsoft and Amazon.
But candidly, the core reason that companies don’t want to work with the federal government and the Department of Defense, in particular, is it’s a business decision. It’s just too damn hard to work with a customer that is going to take you 24 months to get a contract; that contract may then get protested, and there’s no visibility into the process. If you’re a young company you only have so many resources, particularly in your time, that you have to meet milestones for your investors. So you’re going to go work and try to sell it to other industries.
So I think the primary thing the department can do, and this is again, what Mike Brown and DIU are doing a fantastic job at, is how do you make it easier for companies to work with the government so they view it as a trusted and transparent partner? There are plenty of patriots in companies like Google and others that are willing to step up to this mission.
Mr. Muradian: Raj Shah, former head of the Defense Innovation Unit Experimental, and now the CEO of Arceo. An innovative artificial intelligence company.
Raj, it’s always a pleasure. I look forward to talking to you again soon.
Mr. Shah: Thanks, Vago. See ya.
30
