A committee consisting of leaders and scholars from the Massachusetts Institute of Technology (MIT) have released policy briefs that provide a blueprint for governing artificial intelligence (AI) in the U.S. The committee believes that existing regulatory and liability processes can be extended to regulate AI in the U.S. The goal is to strengthen U.S. leadership in the AI sector while reducing potential harm and exploring ways to make AI beneficial to society.
The main policy paper, “A Framework for U.S. AI Governance: Creating a Safe and Thriving AI Sector”, suggests that existing government entities that regulate the relevant domains can also regulate AI tools. The document emphasizes the importance of identifying the purposes of AI tools to enable regulations to be appropriately designed.
The committee proposes that AI providers define the purpose and intent of AI tools in advance so that existing regulations and regulators that are relevant can be assigned. AI providers are also required to set guardrails to prevent misuse and determine to what extent companies or end users could be liable for specific problems.
The policy brief also argues for the creation of new auditing standards for new AI tools, which could be either government-initiated, user-driven, or derived from legal liability proceedings. Steps should also be taken to create a new government-approved “self-regulatory organization” (SRO) specifically for AI, which would have the expertise to interact effectively with the rapidly evolving AI industry.
The policy briefs note that, while existing regulations can cover many aspects of AI, there are areas that will require new legal consideration. The briefs identify copyright and intellectual property rights related to AI, as well as “human plus” legal issues where AI has capacities that exceed human capabilities, as areas needing fresh legal consideration.
The policy briefs also emphasize the importance of promoting research on how to make AI beneficial for society. For instance, the question of whether AI might be used to augment and aid workers rather than replace them is discussed, a scenario that could result in better distributed long-term economic growth.
The committee is composed of a multidisciplinary team that includes economists, computer scientists, political scientists, and electrical engineers among others. The team aims to balance the excitement over AI with concern by advocating for adequate regulation alongside technological advances. They believe that working to serve the nation and the world responsibly requires robust governance and oversight of AI.
