The Problems

Over the last 50 years, the U.S. Code of Federal Regulations has more than doubled in its number of restrictive words, from 400,000 to over 1,000,000. As our economy has become more complex, we’ve layered on additional protections for citizens and the environment. But the growth in regulation means compliance has become extremely costly. Even the most earnest and well-resourced organizations struggle to comply with the routine thousand-page regulation, and American citizens have a paperwork burden of more than 10 billion hours a year.

It has become almost humanly impossible for compliance teams to keep up with regulations in a time-sensitive and cost-effective way. Regulators routinely estimate compliance burdens with their newly released regulations to be in the hundreds of millions of dollars.

In short, there is a regulatory “sludge” problem.

At the same time, increasingly powerful AI needs democratic guardrails as it moves into more autonomous roles. But the choice of human values AI should embody is not a question for a small group of people. It must reflect society’s views on moral questions. This is the societal-AI alignment problem.

Turning regulations into AI agents is the solution to both of these critical issues.

The Solution

We are solving the sludge problem by converting regulations into computer programs that can make autonomous compliance determinations. This enables compliance teams to delegate tasks to Regulatory AI via high-level instructions, while at the same time validating the resulting work product at any level of granularity. Norm Ai agents serve as a round of context-aware compliance checks that pinpoint what may be problematic with what regulation, enabling compliance teams to quickly understand the legal basis for a determination. Businesses can move faster, do more, and be more compliant, all at the same time. This dynamic flips the sludge problem on its head, potentially even unlocking the ability of the government to regulate more at lower cost to society.

Today, Norm Ai agents primarily work alongside compliance teams to deliver proactive compliance. Soon, interacting with AI agents will be a part of daily life, presenting enormous governance challenges and incredible opportunities. Norm Ai agents will work alongside these other AI agents who have been entrusted to automate business processes. Here, the role of the Norm Ai agent is to automatically ensure that actions other AI agents take are in compliance with law. This enables their deployment. Because public policies reflect our society’s collective will, Norm Ai agents will help solve the societal-AI alignment problem through this oversight.

I founded Norm Ai after more than a decade of research experience at the intersection of AI and law, most recently at Stanford’s Center for Legal Informatics. I was also the founding CEO of Brooklyn Investment Group, an AI-powered investment software platform and SEC Registered Investment Adviser, where I now serve as Chairman. My research experience informs the Norm Ai approach to theoretically and empirically grounded AI product development, while my prior experience leading an AI technology company in a highly regulated environment motivates our focus on automating corporate compliance.

Our team brings together the unique blend of AI, legal, and engineering expertise that is required to solve the deeply interconnected regulatory sludge and AI alignment problems by building Regulatory AI. If you would like to contribute toward our mission, you can find our job openings here.

You can read more about Norm Ai in this recent article from Bloomberg Law.