Nobody's Ready for the Killer Robot
It was another busy year for everybody's favorite automotive-industry disruptor, space-travel visionary and potential James Bond villain Elon Musk. Tesla surpassed Ford and General Motors in market capitalization; the Gigafactory began churning out lithium-ion batteries; his neighborhood roofing company began installing solar panels that aren't crimes against architecture; he's sending two rockets to Mars; he started digging a giant tunnel under Los Angeles; and he dissed President Donald Trump over the Paris Climate Accord. (OK, he had a few misses too; just ask anybody on the Model 3 waiting list.)
Given all this, you may have overlooked another of Musk's 2017 initiatives: saving humanity. Last summer, he and a bunch of other tech-industry A-listers -- including Google's artificial-intelligence guru Mustafa Suleyman -- wrote a letter to the United Nations urging a ban on killer robots. The future dystopia they anticipate would make you nostalgic for Skynet:
Lethal autonomous weapons threaten to become the third revolution in warfare. Once developed, they will permit armed conflict to be fought at a scale greater than ever, and at timescales faster than humans can comprehend. These can be weapons of terror, weapons that despots and terrorists use against innocent populations, and weapons hacked to behave in undesirable ways. We do not have long to act. Once this Pandora’s Box is opened, it will be hard to close.
Meanwhile, the Pentagon was also writing a letter -- its 2018 budget request asking Congress for $13 billion in science and technology money and another $10 billion for space-based systems. With all respect to Mr. Hyperloop, he's going to find the military-industrial complex a much tougher foe than the dinosaurs of Detroit.
Nonetheless, as somebody who's a tad nervous to be alone in the house with Alexa, I'd like to think there are folks involved with autonomous weaponry who share Musk's concerns, if not his tired mythological metaphors. So this week I found one such person: Robert H. Latiff, the author of "Future War: Preparing for the New Global Battlefield."
Latiff retired from the Air Force as a major general in 2006, last serving as the director of advanced systems and technology at the National Reconnaissance Office. Since then, he has worked in the defense industry and as a lecturer at Notre Dame and George Mason universities. Here is a lightly edited transcript of our discussion:
Tobin Harshaw: General, you left the military more than a decade ago, so why come out with this book now? Did your experiences in the private sector and academia inform your view of the "new global battlefield"?
Robert Latiff: Well, I'd actually been thinking of this since back around the time of the invasion of Iraq. When I saw some of the things that were going on, not the least of which was Abu Ghraib, they bothered me a lot. Then I went to work, as you might expect, as a defense contractor, and it was not a bad job. But neither there nor while I was in the military did I actually hear anyone ask whether we should be doing some of the research we were doing. You know, some of it was a little scary -- I don't know that it was necessarily unethical -- but nobody ever asked the question.
TH: Can you give an example or two?
RL: Generally, some of the things that had to do with biology were a little frightening, things like synthetic biology where you don't really know the ultimate implications. And some of the work with electromagnetics was a little scary, particularly as it had to do with humans and lethality.
TH: Got it. So you spent some time in the private sector, and …
RL: When I finally left and began teaching and doing consulting I had some time and these things were still bothering me, and I contacted the people at Notre Dame. They jumped at the chance of having me put together a course, which became hugely popular. Ultimately I was asked to write a book nominally but not totally based on the course I was teaching.
TH: And how did the students at Notre Dame react to the material?
RL: First of all it frightened them a little bit. I think they were probably more surprised than anything else. Many of them had never had any exposure to the military at all.
When you start talking to them about some of the newer weapons, it kind of blew their minds. For many of them the first response was, “Yeah we've got to do this.” But when I started asking them to think about the ethical implications, they sort of stepped back from that, and I think they really got a lot out of it.
TH: As long as there's been warfare, there have been arguments about the ethics of warfare. Thomas Aquinas's “just war” theory is probably the most famous. As we look at today's technologies, which ones raise the biggest questions for you viewed in that long tradition?
RL: I think that artificial intelligence and autonomy raise probably the most questions, and that is largely because humans are not involved. So if you go back to Aquinas and to St. Augustine, they talk about things like "right intention." Does the person who is doing the killing have right intention? Is he even authorized to do it? Are we doing things to protect the innocent? Are we doing things to prevent unnecessary suffering? And with autonomy and artificial intelligence, I don't believe there's anybody even in the business who can actually demonstrate that we can trust that those systems are doing what they should be doing.
TH: Well, we know that a lot of people are worried about this. Last year a bunch of tech-industry bigshots wrote a letter to the U.N. urging a ban on killer robots. There was a group that held a meeting in November, under the authority of the Convention on Certain Conventional Weapons that tried to get the ball rolling on a global ban. To me, this honestly sounds like a lot of do-gooder nonsense. But do you think an international effort can be made to limit the sort of apocalyptic level of things that people worry about?
RL: I do think that there is, and there should be. Incidentally, the people who are supporting this weapons ban actually contacted me at one point. And my response to them was that I don't think it's do-gooder nonsense. I think it's the right intention but it's the wrong approach. And let me explain. First, I don't think bans ever work. And second, I don't believe developed nations are going to participate in a ban. And so whenever you ban something, pretty much it just goes underground and you can't police it because in international law there’s no policing.
I am a big fan of arms-control agreements and nonproliferation agreements. Countries like the U.S. and Russia and China, they're going to do this regardless. And if we could create some kind of arms-control agreement where we could maybe limit the things that we do collectively and have some kind of verification regime -- and more importantly, agree to try to at least contain the proliferation -- that would be a step forward. Yes, places like North Korea and other rogue nations and ISIS, they're ultimately going to get some of this stuff. But it would be a lot easier to police and verify an arms-control agreement than a ban.
TH: I'm also a big fan of nuclear nonproliferation agreements. But with that, we're talking about major hardware and a vast industrial base that you need to build the stuff, and then actual nuclear material that is hard to produce and hard to hide. Whereas with AI and cyber weapons, production is pretty easy to hide. How do you do verification?
RL: Well, that is the question of the day, and I don't have any pithy answers, other than to the extent that we can create some sort of verification agreements. The major powers are pretty good at that. With today's agreements, we have overflights of Russia and they'll overfly us, and we'll take satellites over there and they'll bring satellites to look at us.
So with new technologies, we'll need to get some opportunity to just go into these laboratories and see what they're doing, and we're pretty good at projecting what their real capabilities are. I don't know exactly what a regime would look like, but it would clearly be better than just having a ban and trying to find the needle in the haystack.
TH: A few months ago I interviewed former Deputy Defense Secretary Bob Work, who oversaw the Barack Obama administration's cutting-edge Pentagon modernization. He calls this initiative the "third offset," and says it is necessary because our great-power competitors, Russia and China, are achieving parity with our "second-offset" guided munitions and integrated battle networks. Do you view the challenge of the future in that same way?
RL: I do and I agree with the secretary that they really caught up with us. Just look at the recent demonstration of Russian cruise missiles in Syria -- those were pretty good cruise missiles. With things like autonomy and artificial intelligence and hypersonic weapons and electromagnetic weapons, if they have not achieved parity, they're coming up really quite fast. So we won't have the third offset for nearly as long as we had the second offset.
TH: Right -- the advantages of the first two offsets, first in nuclear technology and then in precision weapons, lasted decades. This one is probably more short- to medium-term. So, given that, which of these new technologies should the Pentagon be focused on?
RL: I think we still have an advantage in autonomy and cyber. I think the one that worries me more than any of the others, and it isn't clear to me that we actually have an advantage, is electronic warfare. I'm not necessarily talking just about huge nuclear electromagnetic pulses, I'm talking about everything from very small electronic warfare to great big electronic warfare.
TH: Such as jamming systems?
RL: Well, jamming systems is one. Or electronic pulses that could either destroy or interfere with some of our electronic systems. It's a little bit like cyberwarfare, but it's using electromagnetic pulses. And then battlefield weapons that incorporate microwaves and things like directed energy beams.
TH: So, not just killer robots, but also "Star Trek" phasers. Speaking of which, we know that outer space is, under U.N. treaty, supposed to remain unweaponized. But I think it's on the verge of getting weaponized pretty quickly. China, Russia and even countries like India are putting a lot of money into space. Do you think we need a space nonproliferation effort?
RL: I would personally welcome a space nonproliferation effort. Again, I think the major powers are probably not anywhere near wanting to do that. When you say "Star Wars" or "Star Trek," I don't worry so much about that -- battles going on in space. I even don't worry about the stationing of weapons in space that might have an effect on the Earth. If you know anything about physics and orbital mechanics, that's just way too expensive.
What I do worry about is that the U.S. -- and increasingly China and Russia -- is extraordinarily dependent on space systems. Everybody knows that. And so a ground-based anti-satellite system, or lasers or electromagnetics that might interfere with the functioning of our very critical space systems, or even on-orbit systems that might interfere with or ultimately perhaps destroy one of our satellites -- these are all extremely worrisome.
TH: What else keeps you up at night?
RL: The whole approach that the DoD is taking to autonomy worries me a lot. I’ll explain: They came out with a policy in 2012 that a real human always has to be in the loop. Which was good. I am very much against lethal autonomy. But unlike most of these policies, there was never any implementing guidance. There was never any follow-up. A Defense Science Board report came out recently that didn't make any recommendations on lethal autonomy. In all, they are unusually quiet about this. And frankly, I think that's because any thinking person recognizes that autonomy is going to sneak up on us, and whether we agree that it's happening or not, it will be happening. I kind of view it as a head-in-the-sand approach to the policies surrounding lethal autonomous weapons, and it cries out for some clarification.
To contact the editor responsible for this story:
James Gibney at email@example.com