Today, I’m sharing my conversation with Divya Siddarth, Co-Founder and Executive Director of the Collective Intelligence Project (CIP) about how we might democratize the development and governance of AI. We discuss:
* The CIP’s work on alignment assemblies with Anthropic and OpenAI — what they’ve learned, and why in the world a company would agree to increasing public participation.
* The #1 risk of AI as ranked by the public. (Sneak peek: it has nothing to do with rogue robots.)
* Are participatory processes good enough to bind companies to the decisions they generate?
* How we need to fundamentally change our conception of ‘AI expertise.’
* How worker and public participation can shift the short-term thinking and incentives driving corporate America.
* Should AI companies become direct democracies or representative ones?
* How Divya would structure public participation if she had a blank sheet of paper and if AI companies had to adopt the recommendations.
That’s it for now,
Charley
This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit untangled.substack.com