Jensen Huang: Using AI to Combat AI Misuse

At a recent event hosted by the Bipartisan Policy Center in Washington, D.C., Nvidia CEO Jensen Huang shared his thoughts on the future of artificial intelligence (AI) and its potential to combat the darker sides of its own technology. He asserted that “it’s going to take AI to catch the darker side of AI,” emphasizing that as AI generates fake data and misinformation at unprecedented speeds, it will require equally advanced AI tools to detect and mitigate these threats.

Huang drew parallels between AI’s evolution and the ongoing challenges in cybersecurity, noting that nearly every company faces the risk of cyberattacks at any given time. “It’s going to take even better cybersecurity to defend ourselves,” he stated, highlighting the necessity of staying ahead of potential threats with enhanced AI capabilities.

Public Concerns Over AI and Elections

As the U.S. approaches federal elections in November, concerns surrounding AI-generated misinformation have intensified. A Pew Research Center survey indicated that nearly 60% of Americans are “extremely” or “very” worried about AI’s role in creating fake information about presidential candidates. The survey revealed a stark division in sentiment, with 40% believing AI will be used “mostly for bad” during the elections.

Adding to these concerns, an unidentified official from the Office of the Director of National Intelligence recently warned that foreign actors, including Russia and Iran, may utilize AI to manipulate election-related content, such as altering videos of speeches.

Huang urged the U.S. government to actively engage as an “AI practitioner,” rather than merely regulating the technology. He suggested that every government department, particularly the Energy and Defense sectors, should adopt AI practices, and even proposed the creation of a U.S. AI supercomputer to drive innovation.

“The scientists would be more than happy to jump on it and create new AI algorithms to advance our country,” he noted.

The Energy Demands of AI

Looking to the future, Huang discussed the increasing energy demands that AI technologies will impose. He predicted that future AI data centers may require “10 times, 20 times” more energy than current facilities, which already account for approximately 1.5% of global electricity consumption, according to the International Energy Agency.

Huang proposed that AI data centers could be strategically located in areas with excess energy that is difficult to transport, allowing for efficient use of resources. “We can build a data center near where there’s excess energy and use the energy there,” he suggested.

Jensen Huang’s insights highlight the dual-edged nature of AI: while it offers incredible potential, it also poses significant risks, particularly in the realm of misinformation. His call for a proactive, engaged government role and the need for substantial energy resources reflects the complexities of navigating an increasingly AI-driven future. As AI continues to evolve, the responsibility to harness it for positive outcomes while mitigating its risks remains a critical challenge.