In 2017, Shield AI’s quadcopter, with no pilot and no flight plan, could clear a building and outpace human warfighters by almost five minutes. This was evidence that autonomous robots can help protect civilian and military lives. It was also proof that Shield AI, a startup just two years old, could ask the US government for a large-million contract for a system for coordinated, exploratory robots. Or would they be scared away?
Harvard Business School Professor Mitch Weiss , and Brandon Tseng , Shield AI’s CGO, discuss the challenges entrepreneurs face working with the public sector and how investing can help entrepreneurs and governments work together to solve large problems in the case, “Shield AI. “
BRIAN KENNY: In the 1983 film WarGames, Matthew Broderick plays a 17 year old computer whiz who unwittingly hacks into the US Military Defense System and triggers a global nuclear war simulation with a self-learning computer. This film is one of many that explores the dangers of military artificial intelligence. This theme is a strong one. A 2017 survey by the Guardian found that 72% of Americans fear that robots will take over our lives. While the notion of robot killers is appealing, AI in the military’s main focus is on providing robot assistance on the battlefield and getting robots to perform tasks that are too dangerous or tedious for humans. It seems obvious that technology can protect soldiers on the battlefield. But at what cost? Today, we will discuss Shield AI with Mitch Weiss, case author, and Brandon Tseng, case protagonist. Brian Kenny is your host. You’re listening to Cold Contact over at HBR Presents Network. Mitch Weiss is a researcher on digital transformation and innovation ecosystems. He is the author of We The Possibility: Harnessing public entrepreneurialism to Solve Our Most Urgent Issues, a book on public entrepreneurship. Shield AI’s Chief Operating Officer and co-founder is Brandon Tseng. He is also a Navy Seal and Harvard MBA. I am grateful to you both for being here today. It was great to have both of you here.
MITCH WeISS: Brian, thank you.
BRANDON TENG: We are grateful. It’s great to be here, ex-Navy Seal.
BRIANKENNY: I figured once I was a Navy Seal. It doesn’t get better.
BRANDON TENG: It’s kind of like that. Yes.
BRIAN KENY: We love having Brandon as the main character on the call. Because we want to hear from someone who has lived the case and can give us their perspective. Mitch, this case was very interesting. This case touches on many important topics and I believe artificial intelligence is a topic that people are often interested in. It was a bit of a tease in the introduction. They may not fully understand it, and there is a lot of uncertainty as to what it is and its intentions. People will benefit greatly from learning about Shield AI. I would like to ask Mitch to set the scene and tell us your cold call when you enter the classroom.
MITCH WeISS: Two years have passed since Brandon Tseng and co-founders founded Shield AI in San Diego. The quad copter, which is AI-enabled, can self-navigate buildings. In some tests, the flying robot was able to outperform human war fighters by up to five minutes. It can save civilian and military lives. The US President said that we need AI to defend, and both the sector of defense as well as the past sector defense also agreed. They have even traveled to Silicon Valley and other places to make this happen. And no wonder why, because the President of Russia of Vladimir Putin has also said, “Whoever wins AI will win the world.” And the Chinese President has said, “China must firmly seize the initiative in the race for AI.” So tell me this, the US Department of Defense spends well more than $600 billion a year why aren’t they showing up at Shield AI with dump trucks full of money?
BRIAN KENNY : That’s a great question. It’s a dramatic opening question. You must get the room moving.
MITCH WeISS: This gets the room moving, the topic and the case, and the protagonist. Yes.
BRIANKENNY: I know that you and me have had a lot of fun talking about the cases you have written. We had a conversation about a case a few years back about Toronto and its smart city. Many of the concerns about the creepiness of AI were raised in that case. These same questions can also surface here, I believe. Could you tell us about the motivation behind writing the case? How does it relate to the types of things you consider as both a scholar and teacher?
MITCH WeISS: You mentioned that you spent a lot of time worrying about whether big public problems can be solved. This potential answer has been something I’ve been researching. It is possible to solve public problems if you are more creative. Although new ideas might not work, they could be transformative if we tried them. It is my conviction that if we want to reach the goal of possibility and government, we must have new ideas. These new ideas must be tested and scaled up. It is important to be able to create companies that can sell new ideas and technologies to governments. This was an excellent opportunity to meet Brandon and Shield AI, to learn more about how our companies build themselves to sell government. What are their opportunities? How do they overcome the obstacles they face?
BRIAN KENNY: Awesome. Let me now turn to Brandon. We would love to know more about you beyond your time as a Navy Seal. Please tell us about the journey you took to reach this point.
BRANDON TENG: I have always had a fascination with engineering and technology. However, I also share a strong sense, similar to Mitch’s, of service. I want to be part of solving big problems and to help others. That’s what attracted my to Mitch’s class. After leaving the Navy, I attended HBS, and simultaneously founded Shield AI with my brother, who is also a serial entrepreneur and engineer. It was really about the idea of how we could bring cutting-edge technology, leading edge autonomy, and AI capabilities to a client that needed them very badly, and also build a large organisation, a large company that could marshal resources and marshal teams to make an effect for our customer and the world.
BRIAN KENY: To put a finer point to what Shield AI does. Can you describe the product and how it works?
BRANDON TSENG: Sure. The easiest way to describe Shield AI is that it is self-driving technology used by the US military to fly aircraft. This is the core of our work and if you have ever been in a self-driving vehicle, you will know how magical it is. It takes a lot of engineering and a lot technology to do that. When I show Shield AI’s products, it’s an almost identical feeling. The computer receives sensor data and begins to comprehend what it is seeing, where it is located in the world, and makes decisions about how to navigate it and how to fix problems. This is a simple analogy for a complex technology that doesn’t work in the same way as humans.
BRANDON TENG: Our brain, which processes all of this information, has sensors for our eyes, ears, noses, noses, and taste senses. This is what’s happening in the self-driving tech world with autonomous cars. However, Shield AI also operates at autonomous aircraft.
BRIAN KENNY : In a completely different setting. You know, people think of autonomous cars as Teslas, but you guys operate in a completely different setting. This case describes the idea of the tool being able to enter buildings and sweep them clean. As a Navy Seal, I assume you have some experience. Could you please describe to our listeners what it is like to perform such an exercise? They will be able to see why Shield AI is so important.
BRANDON TSEN: It was very interesting that you started with the Hollywood example. I think many people think of the US military in terms of these Hollywood examples. Then you can move on to the action movie, where the protagonist clears a building and fires a gun in close combat situations. The protagonist is not afraid. Let me tell you, it is terrifying to go into a building and clear it when you know that there are bullets at both ends from either the antagonists or the protagonist. Because it is such an important problem, that is why we chose to first tackle it. It is a very high-stakes problem and I knew how to communicate it to my team of engineers. Shield AI’s first product manger, I helped them understand and empathize.
BRIAN KENNY – This product seems to be something people would love. Mitch, I’d like to return to you for a moment as we consider artificial intelligence in the larger picture. What does that landscape look? Are the United States and the race for supremacy in artificial intelligence where is it?
MITCH WEISS: Perilous. The National Security Council on AI Commission, co-chaired last March by Eric Schmidt and led by many capable Americans, concluded that the US government was not ready to defend the country in the future artificial intelligence era. Here’s the quote. They went on to state that China has the potential, the talent, and the ambition to overtake the US in the next decade, if current trends don’t change. They concluded that we need to change if we want to keep up with the latest technologies. The consequences of not doing so could be catastrophic. There are many things we can do to improve the situation. These include recruiting and training new talent. These technologies require investment. These include demonstrating that there is a model for its use in a democratic society and democratic methods of using it. There are ways we can improve and accelerate our situation. However, I consider it perilous at this time.
BRIAN KENY: Does the US have its R&D, if we concentrate on the Department of Defense? Do they have internal R&D resources that are looking into this? Is it necessary for them not to look outside at companies like Shield AI in order to improve what they are doing?
MITCH WEISS: Well, absolutely. Although we have some of our most talented people working in DARPA and other armed forces under many commands for the Department of Defense, the speed at which our adversaries are investing suggests that we will need to get help both inside and outside of our government to solve this problem.
BRANDON TSEN: Paralysis would be a fair description of the current state of affairs if I could just take some of what Mitch said. I feel for the customer and the government. It’s a difficult problem. It can be difficult to navigate large bureaucracies. I am grateful that people within the government are willing to accept those challenges.
BRIANKENNY: Brandon, how can you begin to address that issue? How did you and your team decide who to talk to and to whom to pitch your idea?
BRANDON TENG: Hard work, trial on error, and talking to many people. Our entrepreneurial journey is very different from many other successful entrepreneurial journeys. It’s great because it gives me hope there will be other companies who can collaborate with the government to solve these problems. Because the government solves the most difficult problems in the world. These problems will not be solved by Shield AI alone. These companies need every support and partnership from the industry in order to solve these problems.