A new research study by a group of researchers from Oxford University has found that Artificial Intelligence (AI) tools used to generate, edit or contextualize social media posts can introduce hidden biases that spread through online networks and shape public opinion.
According to the study by researchers from the Oxford Internet Institute (OII) at the University of Oxford and the Hasso Plattner Institute at the University of Potsdam, large language models (LLMs) often change the direction of social media posts on controversial topics, even when explicitly instructed to keep the original meaning.
The researchers also show, through simulations of real-world social networks, how these small changes could accumulate across millions of interactions and gradually influence broader public opinion.
The findings by the researchers have raised questions about the growing use of AI-powered writing tools on social media platforms and hence has suggested that AI-mediated communication could become a powerful new mechanism for influencing public discourse.
The study, AI-Mediated Communication Can Steer Collective Opinion, authored by Dr. Stratis Tsirtsis, Dr Kai Rawal, Dr Chris Russell, Professor Brent Mittelstadt and Professor Sandra Wachter, has been accepted for presentation at the AI4Good and Technical AI Governance Research workshops at the ongoing International Conference on Machine Learning (ICML 2026) in Seoul, South Korea.
The study also disclosed that biases were similar across different AI systems. It noted that multiple models altered posts in similar directions, favouring some positions such as gun control, marijuana legalization and feminism, while pushing against others such as atheism and the death penalty.
It further highlighted that small changes in individual posts can influence public opinion over time.
The researchers argued that AI systems embedded in social media platforms can shape how opinions spread online, creating new challenges for transparency, accountability and regulation. The study also showed that specific implementation decisions made by platforms can significantly affect the direction and magnitude of AI-generated influence.
The researchers instructed large language models (LLMs) from different providers to transform human-written texts on contested topics into improved social media posts. They then analyzed whether the AI-generated versions systematically changed the position expressed in the original posts.
They used mathematical modelling and computer simulations based on real social network data from X and Facebook to examine how these small changes could spread through online networks and affect broader public opinion over time.
Also Read: AI Agents Can Run Coordinated Political Campaigns: USC Study
