AI Enters Government and Legislative Processes
Massachusetts is launching a ChatGPT-powered assistant for use across its executive branch to improve efficiency. In Washington, the Republican Study Committee has begun using an AI tool called “Byrd Bot” to help draft legislation by automating compliance checks. These moves represent some of the first significant deployments of generative AI within governmental and policymaking functions.
- The Massachusetts contract with OpenAI will cost approximately $4.3 million annually, providing ChatGPT Enterprise licenses to roughly 40,000 employees at a rate of $108 per person per year. The system will operate in a secure, walled-off environment to protect state data, which will not be used to train public AI models. - A core principle of the Massachusetts initiative is a "human at the helm" policy, meaning employees are ultimately responsible for the accuracy and quality of any work produced with AI assistance. This approach mirrors concerns addressed in other government AI rollouts, where human oversight is deemed essential. - The Republican Study Committee's "Byrd Bot" is specifically designed to analyze legislative text to ensure it complies with the Senate's "Byrd Rule." This rule restricts extraneous provisions in budget reconciliation bills, which can pass the Senate with a simple majority rather than the usual 60-vote threshold. - Other states have paved the way for wider AI adoption with successful pilot programs. A year-long trial in Pennsylvania involving 175 employees across 14 agencies found that using ChatGPT Enterprise saved workers an average of 95 minutes per day. - The use of AI-powered chatbots has become widespread in state and local governments, a trend accelerated by the COVID-19 pandemic. By 2021, the crisis had driven at least 36 states to deploy chatbots to handle the surge in public inquiries. - OpenAI is actively expanding its government partnerships through its "OpenAI for Government" initiative. This program provides access to models like ChatGPT Enterprise and the more specialized ChatGPT Gov, which are designed for secure government use. - At the federal level, AI adoption is surging, with the number of reported AI use cases across 11 agencies nearly doubling from 571 to 1,110 between 2023 and 2024. During the same period, the use of generative AI specifically increased ninefold. - The push into government is a key business strategy for AI companies, highlighted by a major OpenAI pilot program with the U.S. Department of Defense. The contract, managed through the Chief Digital and Artificial Intelligence Office, has a spending ceiling of $200 million and focuses on transforming administrative operations, healthcare access for service members, and cybersecurity.
