This year, Professor John Tasioulas, a leading philosopher in ethics and legal philosophy at the University of Oxford, delivered a thought-provoking address on the challenges and opportunities that artificial intelligence presents for democratic societies. Titled ‘Artificial Intelligence and Democracy: From Threat to Promise?’, the lecture explored the transformative impact of AI on governance, autonomy, and participatory democracy.
Professor Tasioulas framed his discussion around three key issues: the overestimation of AI’s capabilities, the growing power of tech corporations, and AI’s potential role in strengthening democratic participation. Throughout his talk, he underscored the need for careful navigation of AI’s integration into society to preserve core democratic values such as autonomy, fairness, and meaningful participation.
The Overestimation of AI’s Capabilities
While some proponents of AI suggest that general artificial intelligence remains out of reach, Professor Tasioulas warned of the risks of overestimating AI’s abilities. AI may mimic human cognition, but it lacks the qualities that define human autonomy and rationality.
“AI may replicate human capabilities, but it risks underestimating the unique aspects of human nature. AI makes mistakes humans wouldn’t, and it lacks rational autonomy. It lacks a kind of engagement with the world—it’s a model of the world, not a participant in it.”
He pointed to the dangers of deploying AI in high-stakes decision-making, such as within the legal system, where reliance on algorithmic reasoning could reinforce biases rather than mitigate them. Without the capacity for true rational reflection, AI remains a tool that, if improperly applied, risks undermining fairness and justice.
The Shift of Power to Corporations
Another major concern raised in the lecture was the growing influence of large technology corporations over democratic governance. In recent years, tech giants have amassed unparalleled power, extending their influence beyond business into societal control and policymaking.
“Recent headlines have seen UK officials suggest treating these companies like nation-states. A shift is occurring, with corporations gaining sovereign-like influence over governance, affecting data privacy and security.”
Tasioulas highlighted the dangers of this shift, where private entities increasingly shape laws and public discourse, often without sufficient democratic oversight. The challenge, he argued, is to ensure that governments retain authority over these corporations rather than cede decision-making power to them.
AI’s Potential for Participatory Democracy
Despite the risks AI presents, Professor Tasioulas offered a hopeful perspective on its potential to enhance democracy. AI, he suggested, could be harnessed to improve citizen engagement and facilitate more inclusive decision-making processes.
“AI could revitalise democracy by scaling democratic deliberation, help identify those directly affected by decisions, facilitate deliberation, and tailor information to individual learning styles, making decision-making more inclusive & participatory.”
By leveraging AI to enhance civic participation, policymakers could create systems that ensure a broader range of voices are heard. AI tools could help structure discussions, counteract misinformation, and make complex policy debates more accessible to the public, fostering deeper democratic engagement.
As AI becomes increasingly embedded in governance and society, Tasioulas urged caution in how we implement these technologies. The goal should not be to allow AI to dictate democratic outcomes but to use it as a tool to strengthen democratic principles. AI can either deepen existing inequalities or serve as a means to empower individuals and institutions. The choice depends on how societies choose to regulate and deploy these technologies.
Watch the full lecture here.