The cost of engaging in cyberspace is much lower than that of human combat.
The same technology that helped a computer beat a human in the ancient Chinese strategic game Go could dramatically change the way wars are fought, a report suggests.
Artificially intelligent systems today can competently identify images and rapidly carry out repetitive virtual tasks. The technology is far from consciousness or replacing human warfighters but even in its current state, it will force the U.S. military to reassess its major strategies, analysts argue.
As adversaries, namely China, invest in AI applications, the Defense Department should preempt any AI-driven attacks by conducting AI-themed war games, and by investing more in its own use of the technology, according to a report written on behalf of the Intelligence Advanced Research Projects Activity, the intelligence community's research and development unit. Harvard's Belfer Center for Science and International Affairs published the study.
Greg Allen, one of the report's authors, recently did a video-based Ask Me Anything, fielding questions posted on Reddit. Here are a few takeaways from that conversation:
The White House has yet to address the military implications of artificial intelligence.
Under Barack Obama, the White House released a series of reports related to AI, including its potential to make some jobs obsolete and raise the possibility that the government should provide a safety net for those whose jobs are replaced. The White House was discussing on a report related to military applications, according to Allen.
Today’s AI is still only very good at very specific tasks. But it’s still poised to transform warfare.
Deep learning technology—a subset of machine learning that builds systems and can teach itself to solve certain problems—has mastered strategic games like chess and Go. Allen calls that “Narrow AI.” It hasn’t yet developed the sci-fi vision of artificial consciousness, and “general AI” would require techniques outside just deep learning. Still, AI systems are pretty good at image recognition, and Defense and the intelligence community are starting to investigate how AI can analyze satellite images, among other very specific applications.
Cyberattacks can be cheaper and quicker than physical military operations, so virtual skirmishes might become more common.
In the cybersecurity realm, there’s “a lot more willingness of states to infiltrate each others’ systems,” Allen said, “essentially to take risks they would not take in the physical domain” because it would be too costly and time-consuming to coordinate. For instance, Russia doesn’t often risk entering U.S. airspace, but has often violated cyber boundaries, he said.
Military bureaucracy could be a huge barrier to adoption.
Some military technology investments go into "stealth technology," traditionally very expensive and complicated so that "nobody else can do it," Allen said. "Contrast that with artificial intelligence, which might be cheap and easy and widely available for commercial technology."
The fact businesses and adversaries are investing in AI could mean everyone has access to the same technological building blocks, which could “make a lot of the stuff that the U.S. military loves about itself,” like “stealth technology,” obsolete, Allen said.