AI models repeatedly write stories about Elias and lighthouses

Oct 19, 2026
Cornell researchers found that major language models often produce similar fictional stories featuring lighthouses, keepers and characters named Elias.

Researchers at Cornell University found that large language models often produce similar fictional stories featuring lighthouses and a keeper named Elias, as detailed in a university news post. Nearly 90% of the stories contained at least one of 11 recurring words, while more than half mentioned a lighthouse.

The researchers generated 20,000 stories using ChatGPT, Claude, Gemini and the Allen Institute for AI's OLMo. The models repeatedly used names such as Elias, Mara and Elara, along with professions including keeper, baker and mayor and settings such as lighthouses and workshops.

The researchers said mode collapse could explain the pattern, with models defaulting to similar themes after training on overlapping datasets. They also found a small group of conversations involving Elias and lighthouses in WildChat, a Cornell dataset containing one million conversations between people and ChatGPT, but have not established a definitive cause.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

AI Brief AI Brief (X)
Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.