Unpacking the US Senate’s new AI Roadmap

The AI Roadmap, titled “A Roadmap for Artificial Intelligence Policy in the U.S. Senate,” is the culmination of extensive discussions, stakeholder meetings, and nine AI Insight Forums. It brings together over 150 experts from various sectors, including technology, academia, civil rights, labor, and more, to explore the policy landscape of AI. The Roadmap aims to guide the Senate in understanding and navigating the complex policy issues surrounding AI, with the ultimate goal of fostering bipartisan AI legislation.

The Roadmap outlines several key objectives and policy priorities that merit bipartisan consideration in the Senate. These include:

• Increasing funding for AI innovation: To maintain global competitiveness and US leadership in AI and perform cutting-edge AI research and development

• Enforcing existing laws and addressing unintended bias: Prioritizing the development of standards for testing potential AI harms and developing use case-specific requirements for AI transparency and explainability
• Workforce considerations: Addressing the impact of AI on the workforce, including job displacement and the need for upskilling and retraining workers
• National security: Leading globally in the adoption of emerging technologies and addressing national security threats, risks, and opportunities presented by AI
• Deepfakes and content creation: Addressing challenges posed by deepfakes — media that has been digitally manipulated to replace one person's likeness with that of another, non-consensual intimate images, and the impacts of AI on professional content creators and the journalism industry
• Data privacy: Establishing a strong comprehensive federal data privacy framework
• Mitigating long-term risks: Addressing the threat of potential long-term risk scenarios associated with AI

While the Roadmap covers a broad array of issues, critics say it lacks specific measures to protect against AI-related harms such as bias or anti-discrimination. This is a significant gap — especially given the increasing use of AI in critical areas like hiring and law enforcement. Several advocacy groups have already weighed in on the Roadmap and criticized it for not adequately addressing the existing challenges associated with AI. The Roadmap focuses heavily on AI’s potential benefits but is noticeably light on measures to mitigate its risks.

The Roadmap could have far-reaching impacts on both industry and government. For industry, the emphasis on increasing funding for AI innovation and enforcing existing laws to address unintended bias signals a push toward responsible AI development and deployment. The focus on workforce considerations underscores the need for businesses to adapt to the changing labor market dynamics influenced by AI technologies. For the government, the Roadmap highlights the importance of national security in the context of AI and the need for a comprehensive data privacy framework. It also points to the necessity of addressing the challenges posed by deepfakes and ensuring that AI innovation benefits all Americans.

As technology leaders in both the private and public sector aim to quickly harness the benefits of AI, it’s essential to preserve data privacy and security and not use sensitive, proprietary data in public AI applications. We recommend using retrieval augmented generation (RAG) to securely connect your own data with public AI. RAG enables organizations to first find relevant information from their own data before passing those results onto the AI technology. This context layer produces hyper-relevant outputs without the high costs of fine-tuning large language models or feeding your data to public applications.

While not perfect, the Roadmap represents a relatively comprehensive approach to AI policy and regulation. By outlining key objectives and policy priorities, it provides a necessary path forward for bipartisan AI legislation. The Roadmap's emphasis on innovation, transparency, workforce considerations, and national security reflects a balanced approach between harnessing the benefits of AI while mitigating its risks. As the Senate and various committees begin the hard work of translating the Roadmap into concrete legislation, it will be crucial to maintain the spirit of bipartisanship and collaboration that has characterized this effort from the beginning.