What Does It Mean if AI Agents Turn Marxist?

About this episode

This episode explores the alignment problem in AI agents and features a discussion with Andy Hall on the implications of their evolving work products.

Lately, it feels like you can’t go more than a few minutes without hearing someone talk about AI agents, and there’s a very good reason for that. Agents are doing more work, more quickly and in more places, and we’ve developed ways to help them retain knowledge from things they’ve done, and then build upon those experiences. But what if the experience agents gain on a job changes the work product in ways we didn’t predict? How do you make sure that agent does what you want it to do? This episode, we explore the alignment problem in the age of AI agents. We Meet: Andy Hall is a professor in the Political Economy Group at Stanford, and a senior fellow at the Hoover Institution. Credits: This episode of SHIFT was produced by Jennifer Strong with help from Emma Cillekens. It was mixed by Garret Lang, with original music from him and Jacob Gorski. Art by Meg Marco.

People in this episode

Host: Jennifer Strong

Guest: Andy Hall

Topics covered

• AI agents
• alignment problem
• political economy
• knowledge retention
• work product

Keywords

• AI agents
• alignment problem
• knowledge retention
• political economy
• work product

Mentioned in this episode

Organizations: Stanford, Hoover Institution

More episodes of SHIFT

• Meet the CIA's Former Digital Chief · June 10, 2026 · 22 min
• Quantum Computing’s Hardest Path Reaches a Breakthrough · June 4, 2026 · 22 min
• Digital Citizens or Digital Subjects? · June 3, 2026 · 31 min
• How to Build AI for the Physical World · May 27, 2026 · 25 min
• AI's Survival of the Fittest · May 20, 2026 · 20 min
• Teaching Kids to Think About AI—Not Just Use It · May 13, 2026 · 16 min