A new kind of political threat may be emerging, but it is far less visible than protests or traditional vote manipulation. Researchers warn that highly realistic personas controlled by AI could soon play a major role in shaping public opinion and influencing democratic systems.
Recent policy forum papers published in science We explain how a large group of AI-generated personas can convincingly imitate human behavior online. These systems can enter digital communities, participate in discussions, and influence perspectives at incredible speed. Unlike previous bot networks, these AI agents can adjust on the fly, respond to feedback, and maintain a consistent narrative across thousands of accounts.
How AI personas imitate real humans online
Rapid advances in large-scale language models and multi-agent systems have enabled a single operator to manage vast networks of AI “voices.” These personas appear authentic, adopt local language and tone, and interact in ways that feel natural to other users.
You can also run millions of small experiments to determine which messages are most persuasive. This allows you to refine your communication strategy and build broad public consensus in real time. In reality, that consensus is artificially created and designed to influence political debate.
Deepfakes and fake news suggest early risks
Although fully developed AI suites are still in the theoretical stage, researchers say there are already warning signs. According to UBC computer scientist Dr Kevin Layton-Brown, these include AI-generated deepfakes and fake news outlets that have influenced recent election conversations in countries such as the US, Taiwan, Indonesia and India.
At the same time, surveillance organizations have identified pro-Kremlin networks disseminating large amounts of online content. This activity is thought to be aimed at shaping the data used to train future AI systems, and could influence how these systems work and what information they prioritize.
Experts warn of growing impact on democracy
Looking ahead, experts believe that AI swarms could have a major impact on the balance of power in democratic societies. Dr Leighton-Brown warned that these systems could change the way people trust online information. “We should not assume that the emergence of these systems will not change society. A possible consequence is a decline in trust in unknown voices on social media, which could empower celebrities and make it harder for grassroots messages to be heard.”
Researchers suggest that upcoming elections could be an important test of the technology. The key challenge is to recognize and respond to these AI-driven influence campaigns before they spread and get out of control.
