US Senators Richard Blumenthal (D) and Josh Hawley (R) have jointly unveiled a comprehensive legislative framework. Their proposal aims to establish a new regulatory body for AI, ensuring that companies undergo a rigorous licensing process before they can engage with advanced language models, such as OpenAI's GPT-4.
This bipartisan framework is a milestone—the first tough, comprehensive legislative blueprint for real, enforceable AI protections. It should put us on a path to addressing the promise & peril AI portends. pic.twitter.com/5CHSfOO9wD
— Richard Blumenthal (@SenBlumenthal) September 8, 2023
Central to the senators' proposal is the requirement for a government license when developing facial recognition and other high-risk AI applications. Companies will be mandated to test their AI models for potential harm, and once these models are deployed, any arising issues must be transparently disclosed. Furthermore, companies should be open to third-party audits of their AI systems.
The initiative of Blumenthal and Hawley follows a more general proposal from senators Elizabeth Warren (D) and Lindsey Graham (R), who suggest the creation of a new federal agency to regulate tech companies such as Amazon, Google, and Meta.
AI licensing has been a topic of discussion for some time now. OpenAI‘s CEO, Sam Altman, has been a vocal advocate, emphasizing the need for AI developer licensing in previous Senate testimonies.
Transparency and Accountability in AI Development
The proposed framework also emphasizes the importance of transparency in AI development. Companies will be obligated to disclose specifics about the training data they use for their AI models. Moreover, if someone is adversely affected by an AI system, they should have a clear legal avenue to challenge the responsible company, a sentiment echoed in discussions intensifying in Washington, DC. Both Blumenthal and Hawley are preparing for a significant event: a Senate subcommittee hearing that will focus on AI. The hearing will delve into the responsibilities of businesses and governments in the realm of AI technologies, with testimonies expected from industry giants like Microsoft's Brad Smith and Nvidia's William Dally.
In addition to their proposal, the senators have worked on another significant piece of legislation: the “No Section 230 Immunity for AI Act”. This legislation is a direct response to the rise of AI-generated “deepfakes” and aims to ensure such content doesn't get a free pass under Section 230 immunity. Senator Blumenthal recently described the bipartisan framework as a pivotal milestone on Twitter, emphasizing that it should guide the nation in balancing the potential benefits and risks of AI.