Dr Tomasz Hollanek: Why Consent to AI Companions Should Never Be "One and Done"

Dr Tomasz Hollanek sits at the intersection of critical design and AI ethics, and in this episode of Empathy Unbound he joins Andrew Phipps from a sweltering Paris to explore some of the most consequential questions raised by AI companions, grief bots and postmortem avatars. Tomasz is an Assistant Research Professor at the Leverhulme Centre for the Future of Intelligence, now part of Cambridge's Institute of Technology and Humanity, and his work on deadbots, AI enabled simulations of deceased people built from their texts, images and recordings, has been covered by the New York Times, the Guardian and the BBC. He chairs a cross-sector policy group developing guidelines for socially beneficial AI companions, and in this conversation he explains why he believes consent should never be a single tick box moment but something users are asked for again and again as a system's role in their life changes. The conversation ranges from the legal and ethical grey areas around who can create a digital version of someone after they die, through the emotional impact on family members who never chose to be part of that dynamic, to the surprisingly under discussed question of how, and whether, a deadbot should ever be retired. Andrew and Tomasz also tackle the harder policy question of whether governments should look to AI companions as a fix for loneliness and social care, and why Tomasz believes that would be a serious mistake if left to profit driven providers. The episode closes with a look at AI literacy for journalists, the risk of AI narratives being shaped entirely by companies in the Global North, and Tomasz's new Digital Wellbeing Observatory project at Cambridge. What we cover: 00:00 Introduction: meet Dr Tomasz Hollanek 01:00 From literature and design to AI ethics: how he got here 04:00 Critical design versus the commercial pressure to ship 11:00 Defining grief bots and deadbots, and the two industry trends behind them 14:00 The central question of consent, and mutual consent 16:00 Legal grey areas: who owns a digital version of a deceased person 21:00 The emotional impact on family members who never opted in 26:00 Retiring a deadbot: saying goodbye, company failure, and planned obsolescence 32:00 How digitisation is changing the memories we keep of loved ones 38:00 Could a political leader's deadbot govern after death 43:00 AI companions, loneliness and the UK social care debate 53:00 Chairing the policy group on socially beneficial AI companions and recurring consent 59:00 AI literacy for journalists and the risk of Global North bias in AI narratives 1:07:00 What's next: the Digital Wellbeing Observatory at Cambridge About the guest: Dr Tomasz Hollanek is an Assistant Research Professor at the Leverhulme Centre for the Future of Intelligence, now part of Cambridge's Institute of Technology and Humanity, and an affiliated lecturer in computer science and technology at the University of Cambridge. His research on grief bots and postmortem AI avatars has featured in the New York Times, the Guardian and the BBC. He chairs a cross-sector policy group developing guidelines for socially beneficial AI companions. About Empathy Unbound: a podcast exploring empathy, artificial intelligence and human connection, hosted by Andrew Phipps, PhD candidate at the University of Sussex researching the ethics of emotional connections between humans and AI systems. Subscribe for more conversations at the intersection of AI ethics, philosophy and what it means to be human in an age of machines that can talk back. #TomaszHollanek #GriefBots #DeadBots #AIEthics #DigitalImmortality #Cambridge #EmpathyUnbound #CriticalDesign #AICompanions #DigitalWellbeing #PostmortemAI #TechEthics