Artificial intelligence is sexy–no doubt about it. From self-driving cars to personal assistant technology that can anticipate your every need, the future of AI looks promising. However, the actual technology can be confusing. And rarely does the reality of AI match up to expectations created by Hollywood portrayals.
In the first of a three part series, the Congressional Subcommittee on Information Technology heard from both industry and academic experts, including representatives from NVIDIA, Intel, Allen Institute for Artificial Intelligence, and the Georgia Institute of Technology.
During last week’s hearing, the Subcommittee, chaired by Rep. Will Hurd, R-Texas, tried to break through the myths and the hype to gain a real understanding of AI, as well its potentials and pitfalls. Based on the hearing, AI has great promise, but it’s grappling with funding and workforce concerns.
The New Space Race?
From the get-go, Hurd makes his views on AI clear–whoever wins the battle in AI innovation wins a strong geopolitical advantage. His position harkens back to space race of the 1960s, when global superpowers were racing against each other to land on the moon. Except, now, it’s about racing to get a computer that can pass the famous Turing Test, which tests a computer’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human. On a more practical level, Feds are looking for AI that can assist intelligence analysts and diagnose an illness before symptoms even appear.
“AI is a technology that transcends borders,” Hurd said. “We have allies and adversaries, both nation states and individual hackers, who are pursuing artificial intelligence with all they have because dominance in AI is a guaranteed leg up in the realm of geopolitics and economics. At the end of this series it is my goal to ensure that we have a clear idea of what it takes for the United States to remain the world leader when it comes to AI.”
Funding Needs
Rep. Robin Kelly, D-Ill., asked the witness how much money the government should invest in AI each year.
The response? Crickets.
After a lengthy pause, Oren Etzioni, CEO of the Allen Institute for Artificial Intelligence, jumped in and explained that he and his fellow witness were not policy or budget experts, but that the investment should be significant.
Etzioni then tied back to Hurd’s previous comments on gaining the geopolitical advantage. He called for Congress to invest “much more” in AI than China’s current investment. While he didn’t know how much China is investing in AI, he said for the United States to remain a leader we needed to invest more.
“Whatever number you come up with, it’s actually 10 times that amount,” said Charles Isbell, senior associate dean, College of Computing at the Georgia Institute of Technology.
The message was clear–staying a leader in AI isn’t going to be cheap.
Training the Future
Rep. Thomas Massie, R-KY, asked when AI would be smart enough to pass the Turing Test. Ian Buck, vice president and general manager of the Tesla Data Center Business at NVIDIA, remarked that worrying about passing the Turing Test was akin to worrying about the overpopulation of Mars–meaning, it was decades off. The witnesses explained that AI was a long-term project and required a future-focused mindset. With that in mind, witnesses stressed the importance of investing in education to train a workforce ready to lead AI innovation.
Isbell stressed that investment needed to start in K-12. Etzioni added that it’s not just about investing in students, but also investing in teacher training to make sure educators are capable of teaching the skills the AI workforce needs.
AI is a complicated technology and one hearing just isn’t enough. The Subcommittee will meet again in March to hear from government agencies on how AI can make government more efficient, effective, and secure.
Rarely do the words “sexy” and “subcommittee hearing” go together–but with AI, anything is possible.
