The Government Harnesses the Power of A.I. while Grappling with Regulatory Challenges

**States Leading the Way in Artificial Intelligence Regulation**
As artificial intelligence (AI) technology continues to rapidly evolve, state lawmakers are taking proactive measures to regulate its use. Instead of immediately imposing restrictions on the private sector, legislators are first focusing on their own state governments, aiming to set a good example while ensuring the protection of citizens from discrimination and other potential harms. This article explores the efforts of various states in developing comprehensive AI legislation and the impact it has on the future of AI regulation.

**States Prioritizing Government Systems**
Connecticut is leading the charge by planning to inventory all of its government systems that utilize AI by the end of 2023, with the intention of making this information accessible to the public. In addition, the state will implement regular reviews of these systems to prevent any unlawful discrimination. State Senator James Maroney, a prominent figure in AI legislation, states that the focus will then shift to regulating the private industry. Maroney plans to collaborate with lawmakers from Colorado, New York, Virginia, and Minnesota to develop model AI legislation that includes broad guardrails, product liability considerations, and impact assessments of AI systems.

**Growing Number of State AI Bills**
A significant number of states, including Puerto Rico and the District of Columbia, have introduced AI bills this year, with 14 states and Puerto Rico already adopting resolutions or enacting legislation. It’s important to note that bills specifically targeting certain AI technologies, such as facial recognition or autonomous cars, are being tracked separately. For instance, states like Texas, North Dakota, West Virginia, and Puerto Rico have established advisory bodies to study and monitor AI systems used by their respective state agencies. Louisiana has formed a new technology and cybersecurity committee to examine AI’s impact on state operations, procurement, and policy.

**Understanding the Use of AI Within State Borders**
The primary objective for state legislators is to gather essential data regarding the usage of AI within their state borders. It involves determining who uses AI and the purposes it serves. Connecticut’s new law was born out of an investigation conducted by the Media Freedom and Information Access Clinic at Yale Law School, which uncovered the extensive, largely unchecked use of AI throughout the state’s government. Similarly, the American Civil Liberties Union of Idaho discovered the undisclosed algorithms that were being used by Idaho to assess individuals with developmental disabilities. The lack of transparency and validation led to the inclusion of corrupt data, raising concerns about the accuracy and fairness of AI systems.

**Challenges and Future Directions**
Different states are approaching AI regulation at varying speeds and with different approaches. Some, like Hawaii, have not pursued any legislation yet due to a lack of clarity on the best course of action. In contrast, states like Hawaii are calling for Congress to adopt safety guidelines for AI usage and impose limitations on its application in law enforcement and the military. However, finding individuals with the necessary expertise to address AI matters within state governments remains a challenge.

**International Developments and the Role of the Federal Government**
While states are making progress, the European Union is leading the way in establishing AI regulations. Discussions on bipartisan AI legislation are ongoing in Congress, with the aim of maximizing the benefits of AI while mitigating potential risks. Ideally, federal regulations would provide the necessary guidance; however, Maroney believes states must take the lead due to the federal government’s limited speed and responsiveness. He points out that data privacy legislation has predominantly originated from state initiatives.

**State-Specific Legislative Efforts**
Several states have proposed narrowly tailored bills to address specific AI-related issues. Massachusetts seeks to impose limitations on mental health providers using AI and prevent dystopian work environments where personal data is compromised. New York proposes restrictions on employers using AI as an automated employment decision tool for filtering job candidates. North Dakota has passed a bill to define the term “person” as not including AI. The state’s governor, Doug Burgum, emphasizes the need for guardrails in AI while also embracing the technology’s potential to streamline state government operations.

**Conclusion**
As artificial intelligence technology continues to advance, state legislatures are taking proactive steps to regulate its use. By prioritizing their own government systems and ensuring accountability and transparency, states are laying the groundwork for comprehensive AI legislation. While challenges remain, states are recognizing the need to address the risks and benefits of AI to protect their citizens and remain at the forefront of technological advancements. The role of the federal government is crucial, but states are leading the way in shaping the future of AI regulation.