Salt Lake City, UT, October 24, 2026 —

An eighth-grader from Salt Lake City, Ellianna Gu, tested three artificial intelligence models—ChatGPT, Claude and Gemini—on how they responded to 84 different teen emotions, according to the information provided about the project.

The description identifies the student, the three models and the number of emotions included in the test. It does not provide the prompts Gu used, how she defined or selected the emotions, or how she evaluated the models’ responses. It also does not report the results or say whether one model performed differently from the others.

Without those details, the test’s findings cannot be independently assessed from the available information. The description does not specify whether the models were given identical prompts, whether responses were scored against set criteria, or whether other people reviewed the answers. It also does not say when the test took place or whether the results were presented in a report, classroom project or another format.

The subject of the test is how AI systems respond to a range of emotions associated with teenagers. However, the available description does not establish what the test was intended to measure or what conclusions, if any, Gu drew. No statements from Gu or representatives of the model developers were included.

As described, the project involved ChatGPT, Claude and Gemini and covered 84 teen emotions. Further information about the testing process and results was not provided, leaving the models’ responses and the basis for any comparison unclear.


Story summarized from the original created by Google News on news.google.com, see more information here.

SignalNewsAI
About The Author