The AI Arms Race

NGA’s Dr. Anthony Vinci speaks at USGIF GEOINTeraction Tuesday


The United States, China, and Russia are “in an arms race for artificial intelligence” (AI), according to Dr. Anthony Vinci, director of plans and programs for the National Geospatial-Intelligence Agency (NGA).

Vinci discussed the importance of preparing for the future in front of a crowd of more than 100 people Nov. 14 at USGIF’s GEOINTeraction Tuesday event, hosted by OGSystems in Chantilly, Va.

Vinci pointed to recent reports in which Russian President Vladimir Putin claimed the leader in AI would be “the ruler of the world.” Meanwhile, he said, China is planning to turn AI into a $150 billion industry by 2030.

“There are these real threats, and they’re not necessarily just the ones we’ve grown used to since 9/11,” Vinci said, emphasizing the importance of dialogue about the future among NGA, the broader Intelligence Community (IC), industry, academia, and organizations such as USGIF.

“There’s a real possibility the U.S. could become second best—that we could lose some of these arms races,” he said. “We have all grown up in a world in which, by far, the U.S. was the dominant GEOINT capability, even before it was called GEOINT. We can’t even imagine a world in which we aren’t, but it’s a possibility, and we need to confront that possibility and ensure it doesn’t happen. We need to remain dominant.”

A graphic visualization of Dr. Vinci’s speech, produced by graphic recorders from OGSystems’ Visioneering team. Click to view full size. Credit: OGSystems

Vinci outlined emerging technologies with the potential to help the U.S. maintain intelligence dominance:

Commercial space: Vinci described the commercial space boom as a “complete game changer” as it opens up the use of new sensors. He also noted the rate at which the technology is advancing, citing a U.K. company that touts real-time, full-motion video from space.

The Internet of Things (IoT): The sheer numbers represent the significance of this technology, Vinci said. According to Gartner, there were 8.4 billion connected things last year, and that number is expected to reach 20.4 billion by 2020. By 2025, the IoT is expected to generate two trillion gigabytes of data per year. “Everything changes for us when we start to talk about that,” Vinci said.

Autonomous Vehicles: “What was one science fiction is now just a reality in our lives and within a few years will be everywhere—and not just in the U.S.,” Vinci said. He added that of particular interest to the GEOINT Community is the fact that all autonomous vehicles are sensors equipped with cameras and LiDAR—and of course the maps that will be needed for the vehicles to navigate. “Commercial industry will go out and map urban areas and well-developed areas,” he said. “But they’re probably not going to map dirt roads in Helmand Province any time soon, and so we need to start thinking about those things.”

Vinci said it’s important the IC confront not just how it will take advantage of these new technologies, but also how adversaries might leverage them.

“We have to prepare for a world where a country like China might try to dominate AI, where terrorists have UAVs and other autonomous vehicles they can use for attacks, where Russia might use IoT devices or other things for spying in our country, and where lots of countries and even non-state actors have access to space,” Vinci said.

In some cases, he added, these things are already happening.

Vinci concluded with a quote from writer William Gibson: “The future is already here—it’s just not evenly distributed.”

Mission Focus: Global Sustainability

The morning program included a keynote and a panel discussion about collaboration between government, industry, and academia to detect and quantify changes and map trends.

IC GEOINT Support to USAF and USSF

Panel highlights how commercial GEOINT contributes to mission success

USAF: From Air to Space

Kenneth Bray reflects on ongoing and future changes for USAF ISR, and how commercial capabilities will help them achieve their new goals.