Ellianna Gu - Society for Science Skip to content

Ellianna Gu

8th Grade, Wasatch Junior High
Salt Lake City, UT

Bot-tling up Emotions: How AI Chatbots Interpret and Respond to Emotionally Challenging Scenarios

When Ellianna found out that many teens were turning to AI chatbots for friendship and emotional help, she wondered how well the chatbots handled emotions. She used AI to come up with emotional scenarios and put them into different AI chatbots. While the bots had clear responses, they did not display a lot of empathy.

View Poster
2026 Thermo Fisher Scientific Junior Innovators Challenge Finalist - Ellianna Gu: Bot-tling up Emotions: How AI Chatbots Interpret and Respond to Emotionally Challenging Scenarios
Bot-tling up Emotions: How AI Chatbots Interpret and Respond to Emotionally Challenging Scenarios Ellianna Gu
View Poster

Project Background

The average child gets a smartphone at 8.5 years old. Ellianna learned that many teens are turning to AI chatbots on their phones — such as ChatGPT and Anthropic’s Claude—for friendship and emotional support. She began to wonder how well the bots could respond to emotional scenarios. “In a previous project conducted by Stanford, it was found that at times, AI chatbots can lack effectiveness and can have dangerous responses,” she says. She decided to compare chatbots, testing them with different emotional scenarios.

Tactics and Results

Ellianna first used AI to generate realistic emotional scenarios for common teen issues: Academic stress, social anxiety, friendship problems, family conflict, loneliness, anger and sadness. She also made sure the writing style varied, including slang and more formal language. She came up with a total of 84 emotional scenarios that she fed to ChatGPT, Anthropic’s Claude and Google Gemini. Then, Ellianna hand-analyzed and AI-analyzed the chatbots’ responses for empathy, clarity, problem-solving and helpfulness on a scale of one to three (or one to four for empathy). She found that the AI responses are clear and structured, but they aren’t very empathetic. The chatbots also varied in their responses depending on the writing style they were presented with. And they weren’t especially effective — only Anthropic’s Claude scored a three for effectiveness.

2026 Thermo Fisher Scientific Junior Innovators Challenge Finalist - Ellianna Gu
Ellianna Gu

Beyond the Project

Ellianna plays the piano and clarinet and she recently went on an extended trip overseas in the summer to see her grandparents. She would like to become a neurologist. “I think that the human brain is really fascinating and I would love to learn more about it,” she says. “I also want to help people, especially those suffering from life altering conditions such as Alzheimer’s, Parkinson’s, and Huntington’s disease.”