Lockheed Taps OpenAI to Solve F-35 Challenges

These are where the contracts are going, specifically in the drone segment in the last two years. But Lockheed is pushing hard on innovation, cost efficiency in terms of military options. This is about the nexus of AI autonomy, air aerospace. The person reshaping that road map is doctor Sarah Hiza, Lockheed senior vice president of technology and strategic innovation. Thank you, doctor Hiza, for being with us and for having us here at Skunk Works. Absolutely. Welcome, Ed, and it's an honor to be here. There's no better place to talk about innovation and technology than Skunk Works based on our legacy, and it's a representation of the center of excellence and center of innovations that we have across Lockheed. And that's the work you're doing now? Absolutely. Right? I I want to get straight to the idea of AI. Sure. Right? What AI actually looks like within Lockheed Mhmm. And and and in terms of the programs for next generation platforms as well. Absolutely. So we have chosen to be, agnostic when it comes to Frontier Labs. That's an intentional part of our strategy. We see it a little bit like a horse race, so we're unclear which model will, play out in the long term. And there are pros and cons to each. Fun fact, we have 55 different large language models that we're using Wow. To run Lockheed Martin. And to your point, we use it for both internal business operations to achieve efficiencies and so forth, whether that's human resources, financial an... Analysis. Yeah. But we also apply it for mission applications. So we have a bunch of fielded systems today, aircraft, missile systems, etcetera. By using AI, we can extend them, make them more reliable, and expand their efficacy. You know, in the news cycle, the concern that's been raised is about an agent misbehaving, and that's been largely, like, flagged in the cybersecurity context. But for the... You know, again, the American public in the world, you know, their fear would be what if an AI agent does something that that does relate to to weapon systems? How do you keep that under control? Well, we do take it seriously, first of all. We've actually been working with AI and machine learning for two decades, and that may raise a question of, like, well, why isn't it more ubiquitous? The log jam was around compute capability. Isn't that amazing? It is. And so we've overcome that in the last three to five years with GPUs, TPUs coming online. But to your point, because we've been working with it for so long, we understand harnessing. But really critical to the way we go about AI is the same way we go about, say, a missile system or an aircraft. We put it through rigorous testing, including corners of the box so that we understand its capability and to ensure the harnessing is sufficient. So in the same way we would go through many hours of flight testing for an aircraft, we do the same thing with AI before we deploy it and then apply it. How is AI changing the the future of warfare? Where where do you think this will manifest most? I think certainly in autonomy, we're gonna see a lot of advancement, specifically initially with crewed, uncrewed teaming, and that gives us a lot of advantage there. And this is kinda what we're talking about with Vectis. Right? The idea that Vectis is part of a mission that could have different types of aircraft? Absolutely. So you could have a piloted, f 35, and it is working in collaboration with the Vectrus in that case. And in another way, it is around interoperability. So right now, we have a lot of individual systems, whether that's Lockheed Martin or our competitive mates within the defense industrial base, and they all don't really talk to each other. So having the right data layer to have interoperability between all of those systems will be enabled by AI. Take me deep into Skunk Works. Behind the wall behind you is somewhere I can't go. It's all classified. But but your teams, the AI work that's actually happening here, what kind of people are those? Where have they come from? Are they from the world of research? Are they from the world of defense? Like, how have you built that up, the the competence? Yeah. And so I mentioned we've worked on this for a couple decades, but you have everything, from, say, software engineers to mechanical engineers. And they've been gaining expertise, and we expose them through testing and training. And so it... You know, I would say, certainly, almost all of them have technical and science or engineering backgrounds, but you don't necessarily have to be a software expert to have become... Found yourself in the middle of AI and applying it in, say, in autonomous systems. Sarah, we've we've talked about how these sort of storied and fabled aircraft were born out of Skunk Works since 1943. Why is Skunk Works the best place for for these leading talents in AI to be working? Why not a different part of of Lockheed or or... Like you do with Fortum, for example, just work with an outside partner to do more of the work there? Yeah. I think that, a lot of the folks who work at Lockheed Martin and here at Skunk Works, we like to see our innovative results integrated but ultimately scaled and scaled at relevance so that we make an impact to the nation and and some cases to the world. And so when you work here, you're able to work on those kind of things that scale of relevance. And you mentioned, like, Fordham. We work with all kinds of different companies. I think that's another really cool part of working at Lockheed Martin is that we work beyond the walls of Skunk Works and beyond the walls of Lockheed Martin. So we work with venture companies. We work with large commercial tech companies. And so it gives us lots of opportunities to see different technologies as well as as well as different types of businesses. There are other, nascent technologies. Nascent, maybe you challenged me on that, but let's take quantum for example. You know, in the in the world, everyone sort of says, when is this moving beyond the lab and research? Think about what's at stake with with Lockheed. Is is quantum actually a useful technology to you to get platforms deployed? This past week, actually, we announced the quantum innovation center, and that is because we think we are on that... The cusp of a next breakthrough around quantum. We're now seeing it, moving outside of the lab. And so for us, yes, we we are interested in quantum compute, and we'll leverage that in the same way like, say, we do with GPUs and TPUs. But specifically for national security, we're looking to use quantum for sensing, remote sensing, secure comms, and position navigation and timing. And those are all really kinda niche to national security. But what's exciting is, to answer your question around the lab, next next year, we plan to do demos, including some from space using quantum technologies. Where is, which piece of emerging technology or where is Skunk Works just ahead? The world underappreciates, but you're already on it. But I I certainly think an autonomy is one of those. Autonomy is not new to us. And then the application of AI, both of those, we've been working on a long time. And because of that, we have the the foundation and the expertise, and we understand the mission so that we can deliver a a holistic solution. Does the customer believe in that that technology and in that expertise? They rely on it every day. You you spend a lot of time, I I I guess, explaining to them the benefit of AI. I don't think we have to explain the benefit. I think they know the benefit of AI, and they have seen it, employed in in systems like Aegis where we have demonstrated that. And, actually, I'll share right now that, we're working a a project. As we advance the f 35, we're bringing on even more advanced sensor capability. And we've discovered some math and physics challenges, and we've actually reached out to open I... OpenAI. We all saw the headline that they were able to overcome the math problem for the the researchers had worked on for years. And, we have some math and physics challenges as we employ this next sensor capability on the f 35, and OpenAI is working next to us to uncover some of those math and physics And is that just that their teams or it's their specific underlying model is the best solution there? Their team is working right beside us. We are absolutely working right beside each other. Embed with Skunk Works. Or Yeah. OpenAI is working with f thirty five team on this math and physics challenge to see if we can uncover, overcome that challenge that we've encountered as we advance our sensors.