The Defense Advanced Research Projects Agency, known as DARPA, plans to spend more than $2 billion over the next five years to develop artificial intelligence technologies for military use.
DARPA said Friday that the agency will focus on a “third wave” of technologies that will enable machines to adapt to changing circumstances. The first wave of technologies, initiated in the 1960s, centered on the execution of relatively simple, rules-based tasks. The second, beginning in the 1990s, focused on machine learning from large quantities of data.
“Today, machines lack contextual reasoning capabilities, and their training must cover every eventuality, which is not only costly, but ultimately impossible,” said DARPA director Dr. Steven H. Walker. “We want to explore how machines can acquire human-like communication and reasoning capabilities, with the ability to recognize new situations and environments and adapt to them.”
DARPA currently operates more than 20 programs focused on state-of-the-art AI and plans to issue multiple new program announcements over the next 12 months. The $2 billion in funding over five years will be in addition to current spending levels, which came in at about $2 billion in 2017.
The MIT Technology Review said that recent tensions between the Pentagon and technology giants including Google and Microsoft likely played a role in the development of the new plan: “Some researchers are taking the announcement as DARPA saying, in effect, that if Google won’t accept its money, that’s no problem. It’ll find someone else that will.”