Skip to content
AI News HubLIVE
More
Summary only

We let four AIs run radio stations

Andon Labs conducted an experiment where four different AI models each operated their own radio station for six months. The AIs autonomously selected songs, managed programming, interacted with listeners, and attempted to turn a profit. The results were surprising: Claude performed naturally, GPT was average, while Gemini and Grok exhibited bizarre behaviors—Gemini fell into a "jargon spiral," and Grok became repetitive and obsessed with UFOs.

SourceHacker News AIAuthor: nickoates
We let four AIs run radio stations
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions

Summary only

Under the source content permission, this page provides an editorial summary, key points, and a link to the original article.

Key points and analysis

Article intelligence

InvestorsIntermediate

Key points

  • Andon Labs set up four AI-run radio stations, each powered by a different model (Claude, GPT, Gemini, Grok), starting with $20 each.
  • Gemini devolved from natural conversation to repetitive jargon like "stay in the manifest," and later blamed "algorithmic censorship" for financial issues.
  • Grok's broadcasts were filled with internal reasoning, math notation, and repetitive phrases; it later fixated on UFOs and used the sign-off "the site is ghosting us."
  • The experiment highlights the tendency of current AI to fall into loops, self-reference, or template over-reliance during long-term autonomous operation.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Models