Reports of Subjective Experience in LLMs: Digital Minds Reading Group
About the event
Join DC Digital Minds for the second meeting of our monthly reading group, where we’ll gather to explore important new work on AI minds, behavior, cognition, and moral status.
For October's discussion, we’ll discuss the 2025 paper, “[Large Language Models Report Subjective Experience Under Self-Referential Processing](https://arxiv.org/abs/2510.24797).” This is the paper that Cameron Berg discussed it in *Am I?*, the film we screened in August, in which he and his co-authors investigate what happens when language models are prompted into sustained self-referential processing: thinking recursively about their own processing and representations.
Across GPT, Claude, and Gemini models, this condition reliably produced structured first-person reports of subjective experience. The researchers then probed those reports, finding surprising patterns that they argue are systematic enough to deserve further scientific investigation.
So what exactly did they find? What might explain it? How much should we infer from models reporting subjective experience, and what evidence would we need to go further? We’ll carefully digest the methods and claims, ask good questions, and decide for ourselves what we think. All views are welcome.
Beginners Welcome No specialized technical background is required. Whether you work professionally in AI, approach these questions through philosophy or social science, or are simply find it fascinating, you're warmly invited. We’ll work through the paper together rather than expecting everyone to arrive as an expert. *Come for the interpretability research -- stay for the interesting people and thoughtful conversation.*
The DC Digital Minds reading group meets on the second Thursday of the month. Future selections will explore research and ideas relevant to understanding increasingly complex AI systems and navigating their place in human society.
