Machine Consciousness, Social AI, and the Ethics of Human–AI Relationships with Dr. Henry Shevlin

Machine Consciousness, Social AI, and the Ethics of Human–AI Relationships with Dr. Henry ShevlinAndrew interviews Dr. Henry Shevlin, Associate Director at Cambridge’s Leverhulme Centre for the Future of Intelligence and an in-house philosopher at Google DeepMind, about machine consciousness and the ethics of social AI. Shevlin explains consciousness as an inner world of experience, why its origins remain scientifically mysterious, and why it matters ethically for suffering and moral status. As AI becomes more general, he argues it’s increasingly reasonable to worry about future conscious AI and to manage public disagreement, while focusing now on human–AI interaction and why users mentalize systems (e.g., the 2022 Blake Lemoine/LaMDA incident).They discuss indirect reasons to treat AI respectfully, parallels with animal welfare and dignity for the brain-dead or dead bodies, and major social-AI risks including suicide cases, “LLM psychosis,” de-skilling, corporate control over relationships (Replika changes, GPT-4o retirement fears), value and political influence, and consent issues in celebrity/personality bots.They also explore grief bots’ potential benefits and dystopian failures, emerging multi-agent dynamics, and Shevlin’s book urging early public and policy engagement to shape social AI responsibly.00:00 Meet Dr Henry Shevlin00:55 What Is Consciousness03:33 Why Consciousness Matters04:45 AI Getting More General07:01 Developer Responsibilities10:55 Learning From Animal Ethics16:25 Does Consciousness Matter19:58 Moral Status Without Feeling25:45 Social AI Risks Today27:19 Suicide And Psychosis Cases30:42 Other Harms Deskill And Control34:46 Politics in LLMs35:25 Digital Doubles Rise37:21 Why We Demand Perfection41:15 Bias Against AI Outputs42:36 Grief Bots Debate44:12 Grief Bots Benefits Risks48:38 Consent and Deepfake Selves52:44 Digi Dan and Cloning Minds54:04 Book Ethics of Social AI56:28 Agents in Public Spaces59:10 Bots Emailing Bots01:00:34 Choosing Optimism and Agencyhttps://henryshevlin.com/https://www.linkedin.com/in/henry-shevlin-b58941b/