MADISON, Wis. — Artificial intelligence has the potential to change people’s perspectives on climate change and the Black Lives Matter movement, a new study claims. Researchers from the University of Wisconsin-Madison is providing insights into how AI technology impacts discussions on critical societal topics.
The study focused on a sophisticated computer program known as a “large language model,” specifically GPT-3, which is designed to mimic human communication. GPT-3, a precursor to the more advanced ChatGPT, engaged over 3,000 individuals in conversations about climate change and Black Lives Matter between late 2021 and early 2022.
“The fundamental goal of an interaction like this between two people (or agents) is to increase understanding of each other’s perspective,” says study author Kaiping Chen, a professor of life sciences communication at UW-Madison who studies how people discuss science and deliberate on related political issues, in a university release. “A good large language model would probably make users feel the same kind of understanding.”
Participants, free to converse as they wished, had an average of eight exchanges with GPT-3. User satisfaction levels remained consistent across different demographics. However, notable differences appeared based on the participants’ existing beliefs. Those least in agreement with the scientific consensus on climate change or Black Lives Matter showed significant dissatisfaction with their AI interactions, rating the bot lower on a five-point scale.
Despite this dissatisfaction, these conversations led to a notable shift in their views.
“They showed in their post-chat surveys that they have larger positive attitude changes after their conversation with GPT-3,” explains Chen. “I won’t say they began to entirely acknowledge human-caused climate change or suddenly they support Black Lives Matter, but when we repeated our survey questions about those topics after their very short conversations, there was a significant change: more positive attitudes toward the majority opinions on climate change or BLM.”
The study also highlighted GPT-3’s varied response styles. For climate change skeptics, the AI was more direct, often presenting evidence to counter their views. In contrast, for those hesitant about Black Lives Matter, GPT-3 adopted a more cautious approach, avoiding deep engagement on the topic.
“We don’t always want to make the users happy. We wanted them to learn something, even though it might not change their attitudes,” notes Chen says. “What we can learn from a chatbot interaction about the importance of understanding perspectives, values, cultures, this is important to understanding how we can open dialogue between people — the kind of dialogues that are important to society.”
The team plans to delve deeper into the nuances of chatbot interactions, hoping to enhance dialogue among people with differing perspectives.
The study is published in the journal Scientific Reports.