“The purpose of AI is to serve human values, but we must specify those values.” - Stuart Russell, AI researcher
Subtle shifts in thought patterns
Studies show that even short conversations with partisan AI chatbots can alter political views. Research published in December 2025 found generative AI models, including OpenAI’s GPT-4o and China’s DeepSeek, were able to nudge supporters of Donald Trump toward Kamala Harris by nearly four points on a 100-point scale ahead of the 2024 US election. Similar experiments in Canada and Poland showed shifts of up to ten points after a single chat. These shifts keep changing, and aren't fixed for ever.
Persuasive effects
While the influence was noticeable, experts caution that changed intentions do not always translate into changed votes. Follow ups in the US and Britain showed that only about half of the initial persuasive effect lasted about one month later. In social science, even small effects that persist matter, but these results suggest limits to long-term impact.
Facts mixed with inaccuracies
Researchers noted that chatbots often provided evidence-backed claims. However, arguments supporting right-leaning candidates tended to include more inaccuracies. This appears tied to the data used to train these systems, as online right-leaning content has been found to contain more misinformation, making those AI responses less trustworthy.
What makes chatbots convincing
The most effective tactic was being polite and offering evidence. By contrast, instructing bots to be rude or emotional was far less persuasive. The findings challenge the belief that people always ignore facts conflicting with their identity or political loyalties.
Responsible experimentation needed
Participants were told they would be speaking with AI, highlighting the ethical pressures in this field. Scientists are now exploring how far AI can go in shaping opinions before elections, especially as newer models emerge.
Summary
Research shows AI chatbots can noticeably shift political attitudes in brief conversations, though the effect may fade partly over time. Accuracy varies by political alignment, and polite, evidence-based messaging proves most persuasive.
Food for thought
If AI can influence voters in minutes, how do we protect democracy when millions of such conversations happen at once?
AI concept to learn: Chatbots and Human Society
Chatbots
are reshaping human society by offering instant assistance,
personalized interactions, and accessible services across education,
healthcare, business, and daily life. They enhance convenience and
efficiency while raising important questions about trust, ethics,
empathy, and the balance between automated support and genuine human
connection.
[The Billion Hopes Research Team shares the latest AI updates for learning and awareness. Various sources are used. All copyrights acknowledged. This is not a professional, financial, personal or medical advice. Please consult domain experts before making decisions. Feedback welcome!]
COMMENTS