This blog is the chronicle of a personal impact project created as part of the CAIS AI Safety and Ethics course. The idea is to encourage the federal legislature to take action on AI Safety by contacting them directly, explaining the problem, and identifying actions they can take.

AI will have a huge impact on the world over the next few years and decades. It’s developed more rapidly than many had expected, it’s already being deployed in the real world, and it’s already having widespread consequences, both good and bad. As it continues to improve, it’s important that we take steps to maximize the benefits and minimize the harms.

I am targeting the federal government for a few reasons:

  • Because government involvement can coordinate private-sector actors and discourage the worst abuses. Private companies have incentives problems, and while government is not immune, it at least has different incentives from private actors that will help to constrain companies from the worse excesses.
  • Because I am an attorney, and while I have some background in technology, it’s not nearly enough to directly contribute to technological fixes. Legal interventions are my comparative advantage.
  • Not least, because I live in Washington DC. Congress is my neighbor, and I can show up in person at Congressional offices in an age when even a voice on the telephone is increasingly suspect.

I’m also blogging the process here so that others might build off of it. Part of what lead to this project was a discussion on the the Baysian Conspiracy podcast with Dave Kasten, where he suggested people could have a larger-than expected impact on policy outcomes by just contacting policy-makers. That small push was the impetus I needed to undertake this project. So I’m offering my experiences here in the hope it will give someone else the small push they need.