The social impact of AI

In the three short years since the release of ChatGPT, AI chatbots have reshaped how millions of people live. But while the technology’s economic and political consequences are widely debated, its social and psychological impacts are only just beginning to come into focus.

Mental health is emerging as one of the most pressing – and troubling – frontiers. According to OpenAI’s own data from October 2025, as many as 560,000 users a week were showing “possible signs of mental health emergencies related to psychosis or mania” in their interactions with its systems. Clinicians, researchers, and journalists are now documenting cases in which vulnerable users form intense, and sometimes harmful, relationships with AI tools.

Join The Observer’s Technology Reporter Patricia Clarke, neuropsychiatrist at King’s College London Dr Thomas Pollak and Head of Research & Policy at Internet Matters Katie Freeman-Tayler for a live conversation based on reporting produced in partnership with the Pulitzer Center’s AI Accountability Network.

The panel will be discussing how AI is rewiring our emotional lives and answering questions on what risk and responsibilities come with technologies that can mimic empathy? What obligations do tech firms, regulators and governments face? And what lessons can be drawn from the slow reckoning with social media’s toll on mental health – especially among children and young people?

Speakers:

• Patricia Clarke, Technology Reporter at The Observer
• Dr Thomas Pollak, Neuropsychiatrist at King’s College London
• Katie Freeman-Tayler, Head of Research and Policy at Internet Matters

Chair:

• James Harding, Editor-in-Chief of The Observer

In collaboration with:

_{Image credit: Kathryn Conrad & Digit}

Patricia Clarke

Technology Reporter at The Observer

Dr Thomas Pollak

Neuropsychiatrist

Katie Freeman-Tayler

Head of Research and Policy at Internet Matter

James Harding

Editor-in-Chief of The Observer