How to Build a Better Great Wall

In my last column, I touched on the fact that much of the government was still shutdown due to the impasse over border wall funding. Truly, I thought that it would all be resolved by now. But with no end in sight, I wanted to talk about using technology—specifically artificial intelligence and drones—to create an impenetrable defense along our borders that would be mostly invisible and yet highly effective. Much of the technology needed to support such an endeavor already exists today, and such a system might just be acceptable to both sides of the great wall debate.

First off, it’s important to mention that walls by themselves have never been effective at stopping humans from moving through, over, under or around them. The most famous wall ever built, The Great Wall of China, did not prevent Mongol raids into Chinese territory, or later Han Chinese migration into Manchuria, in sections where the wall was unmanned. During the Ming dynasty, over 25,000 watchtowers were added to the wall, along with thousands of troops to man them. There were even army units stationed nearby, who could respond to signal fires and other alerts sent from the watchmen. Only then did the wall become an effective counter to both migration and invasion, though at an incredible cost in both materials and manpower.

But let’s look at walls on a smaller scale, say in prisons. Obviously, the point of a prison wall is to keep those convicted of crimes inside. But how effective do you think they would be if we sent all the guards home, and just let the inmates walk around the prison yard unsupervised? I would predict that within a couple days at most, the prison would be empty.

People are extremely clever, and even the dullest human can outthink a static defense. Give a person unlimited time, and eventually they will find a way to destroy or circumvent any wall that just sits there. As such, the problem has really always been, going all the way back to the fourteenth century, one of manpower.

While it’s not feasible to take the ancient Chinese approach of building thousands of manned watchtowers backed up by army units, we do have the technology right now that could lock down the entire border with or without a wall. There would be two main components in such a system, drones to run the patrols and artificial intelligence to process everything that the drones are seeing and detecting. There would also need to be human border guards, but if the first two components are working properly, then humans would only be responding to locations where they are needed, not wasting their time on patrolling and not responding to false alarms.

The first component, the drones themselves, are practically an established technology at this point. Drones have already proved highly effective in public safety roles, searching for fires or lost people, and netting quite a few high-profile success stories. At the recent Consumer Electronics Show, there were many new drone models offering long flight times, rugged durability and some level of artificial intelligence. One company is even starting to sell personal home security drones that will automatically patrol your property.

So drones would not be a problem, though they may require a series of base stations be built where they could periodically recharge. The trickier part would be providing artificial intelligence to process everything that the drones are finding on their patrols, and also to fly them so that teams of human pilots would be unnecessary. Here too, a lot of work has already been completed, at least on a smaller scale.

Several years ago I was shown a demonstration of artificial intelligence being applied to surveillance camera feeds. The AI was able to determine, for example, if someone was behaving outside of the normally seen behavior for that area. That might mean they were walking in the wrong direction down a one-way hallway, or milling about while everyone else was moving. It could also scan for things like weapon profiles, and if attached to a database, could use facial recognition to search for both the identity of the person being watched and to check to see if they had any outstanding warrants. That technology has already been used to protect things like political conventions and sporting events, and is rumored to have been heavily leveraged inside New York City’s Domain Awareness System since it launched in 2014.

Another critical component to such a system would be protecting it from hacking or manipulation, and ensuring that the feeds the drones and other devices are submitting are valid, and have not been tampered with in any way. For that, the Homeland Security Department wants to see if blockchain can be used. Homeland Security recently awarded a company called Factom a $192,000 grant to test blockchain’s ability to protect surveillance feeds in a real-world environment alongside Customs and Border Protection agents.

At some point, humans would need to be brought into the border protection loop, but a wall of intelligent drones could ensure that nobody could illegally enter the country without being spotted, and those responding officers would know exactly what kind of situation they are facing each time. And because the drones could patrol up to 20 miles or more into the interior, nobody would be able to tunnel under their protection. For example, if someone suddenly pops up in the middle of nowhere ten miles past the border, then they likely got there using a tunnel, which would trigger an alert.

The idea of creating a robust, intelligent technology barrier as opposed to a static physical one has been floated before. Back in 2017, Rep. Will Hurd, a Republication from a border region in Texas, proposed the Secure Miles with All Resources and Technology Act, which would have deployed “the most practical and effective technology available (such as radar, tunnel detection technology, unmanned aerial vehicles, and sensors) to achieve situational awareness and operational control along the U.S. border.” Although it had some bipartisan support, it never left the subcommittee phase.

Even with a technology wall, perhaps some kind of physical barrier would be useful, if nothing else than to warn people that they are about to enter the United States illegally—and that they will be identified and caught if they proceed. But it could be a picket fence and be just as effective as a stone block. So long as the technology behind the barrier is solid, having a wall or not really doesn’t matter, and is no excuse to keep our government hobbled and shuttered.

John Breeden II is an award-winning journalist and reviewer with over 20 years of experience covering technology. He is the CEO of the Tech Writers Bureau, a group that creates technological thought leadership content for organizations of all sizes. Twitter: @LabGuys