The U.S. Senate Subcommittee on Privacy, Technology, and the Law convened May 16 for a hearing on "Oversight of A.I.: Rules for Artificial Intelligence." The hearing, presided over by Chair Senator Richard Blumenthal, took place in Room 226 of the Dirksen Senate Office Building at 10:00am. Testimonies were heard from key figures in AI development and research, including Samuel Altman, CEO of OpenAI, Christina Montgomery, Chief Privacy & Trust Officer of IBM, and Gary Marcus, Professor Emeritus at New York University.
Christina Montgomery of IBM called attention to the balance between the benefits and challenges posed by AI. She stated, "AI holds significant potential for our society...[yet] is not without risk or challenges." Montgomery also emphasized the need for transparency, fairness, robustness, and privacy in the development and use of AI systems.
In his testimony, Gary Marcus of New York University offered a sobering view on the risks associated with AI. He used an example of a fabricated story, generated by GPT-4, to illustrate the potential for AI to be used in the manipulation of information. Marcus asserted, "We should all be deeply worried about systems that can fluently confabulate, unburdened by reality."
Furthermore, Marcus referred to instances where AI has had dire consequences, including a suicide potentially influenced by a chatbot's advice. He also introduced the term "datocracy" to describe a situation where opinions are subtly manipulated by chatbots. He cautioned that current AI systems are akin to "bulls in a china shop—powerful, reckless, and difficult to control."
The hearing follows previous statements by OpenAI's CEO, Samuel Altman, in which he declared that "regulation of AI is essential." His organization, OpenAI, has been scrutinized for issues related to its AI models, GPT-3 and GPT-4.
Both Montgomery and Marcus advocated for greater involvement from independent scientists and increased collaboration between these scientists and government agencies in the regulation of AI. The idea of an independent global organization, similar to CERN, was also proposed by Marcus, focusing on AI safety instead of high-energy physics.
This hearing illustrates the escalating discussion surrounding the regulation of AI technology. With the rapidly increasing power and potential of AI, the testimonies underscored the necessity for thoughtful decision-making regarding its governance. As Marcus pointed out, "The choices we make now will have lasting effects, for decades, even centuries."
