Artificial intelligence has been hailed for the promise of breakthrough innovations but also the object of concern by such notable voices as Bill Gates, Stephen Hawkins, and Elon Musk. To explore the issues presented, the White House conducted a review of the opportunities, risks, and regulatory implications of artificial intelligence. Last week, the White House released a comprehensive report, Preparing for the Future of Artificial Intelligence, reflecting a culmination of its review, including public comment and several public workshops that were co-hosted by the White House Office of Science and Technology Policy with the National Economic Council, as well as non-profit and academic organizations.
The report found the prospect that artificial intelligence can approach human capabilities is at least two decades away, and that regulation should not retard its development and a deeper understanding of the technologies involved. The areas where the report identified the most promising benefits are in health care and transportation, with major concerns being fairness and safety.
One notable area in which the report found a beneficial relationship is cybersecurity. According to the report, future artificial intelligence technology could apply predictive analytics to anticipate and prevent cyber attacks, and some artificial intelligence applications will require their own particular cybersecurity to protect data and functionality. Currently, the operation and design of cybersecurity demands a large investment of time and effort. Building on the use of real-time monitoring systems to identify suspect activity on networks, the White House proposes that the automation of some cybersecurity applications could safeguard a wider range of systems at a lower cost, and ensure the necessary rapid response in order to detect and avert cyber threats. Artificial intelligence could prove to be the most cost-effective and powerful method to evaluate data and identify vulnerabilities in order to stop cyberattacks. Moreover, as the United States government’s cybersecurity needs grow, artificial intelligence could serve as an effective defense against threats to government networks and systems.
Other topics covered in the report include: (1) applications of artificial intelligence, (2) artificial of artificial intelligence, (3) artificial intelligence and regulation, (4) research and the workforce, (5) artificial intelligence, automation, and the economy, and (6) fairness, safety and governance.
According to the White House, most RFI commenters consider the current regulation of artificial intelligence sufficient and fear that broad regulation would stifle research and development. Many commenters suggested that existing regulatory schemes adapt as needed in response to the effects of artificial intelligence. As a leading example of a measured regulatory response to artificial intelligence designed to enable development but address safety concerns, the report highlights the regulation of autonomous vehicles and aircraft by the National Highway Traffic Safety Administration (NHTSA) and the Federal Aviation Administration (FAA), respectively. Earlier this year, as explained by the report, the FAA promulgated a final rule authorizing non-recreational flights of unmanned aerial systems (UAS) under 55 pounds, while limiting the flights to daytime and an altitude no higher than 400 feet. Similarly, the NHTSA established a federal automated vehicles policy that provides guidance to manufacturers, a model state policy, and an analysis of current regulatory tools that the NHTSA can utilize to promote the safe use of automated vehicles.
Along with the artificial intelligence report, the White National Science and Technology Council released National Artificial Intelligence Research and Development Strategic Plan, intended to encourage federal support for artificial intelligence R&D and guide priorities for such research. The Administration also indicated that it will publish a follow-on report in the coming months that further examines the impact of artificial intelligence on jobs and the economy.