Professor Kate Devlin: AI Companions, Grief Bots, and the Future of Human-AI Relationships

Professor Kate Devlin: AI Companions, Grief Bots, and the Future of Human-AI Relationships

Should you feel guilty for loving your AI companion more than going outside? Professor Kate Devlin, author of "Turned On: Science, Sex and Robots," has spent her career asking why we keep insisting human relationships are the gold standard, and whether that's actually true. Professor Kate Devlin is Professor of AI and Society at King's College London and Director of the Digital Futures Institute. She trained originally in archaeology before moving into computer science, and her research explores how and why people form emotional connections with technology, particularly around intimacy, connection, and ethics. She is the author of "Turned On: Science, Sex and Robots," a contributor to the Oxford University Press volume on AI ethics, a co-investigator on the UKRI Responsible AI UK programme, a board member of the Open Rights Group, a patron of Humanists UK, and a commissioner for the International AI Faith and Civil Society Commission. Kate explains how her background in archaeology gives her a long view of human-technology relationships, observing that while tools change, human fears and hopes remain remarkably consistent across centuries. She is sceptical of AGI as a concept, arguing it is too poorly defined to be useful, & doesn't believe large language models can produce anything resembling sentience. The conversation moves into the emotional bonds people form with AI companions. Kate argues these bonds are genuinely felt on the human side, even when users know full well the AI isn't sentient, in much the same way people form deep attachments to fictional characters or celebrities. She discusses the Pope's recent encyclical on AI, noting its balanced acknowledgement that AI companionship isn't inherently negative, and pushes back gently on the assumption that human-human relationships are automatically the gold standard, given how often human relationships involve conflict, coercion, or even war. Kate traces the history of fembots in fiction back to the myth of Pygmalion, and discusses how AI companion technology, originally built by men for straight men, is now seeing a significant shift toward female users seeking safer, more respectful alternatives to a often-hostile online dating world. She also explains why physical sex robots never became commercially viable, and why a simple app proved far more compelling, a development she compares to the evolution from 1980s adult phone lines to today's AI companions. The discussion turns to some of the most emotionally resonant material in the episode: real reports of grief and heartbreak when Replika removed features from its AI companions, the ethics of "grief bots" that allow people to interact with AI recreations of deceased loved ones, and a striking real-world story from China involving a dying grandmother and an AI recreation of her grandson. Andrew shares his own reflections on his mother's dementia, and what AI companionship might have meant for his mother. Kate also discusses her work on the Responsible AI UK programme, what who holds power in AI development, and why describing AI simply as "a tool" understates its scale and impact. She talks candidly about people who reject AI entirely, often for well-founded reasons around environmental cost, labour exploitation, and creative industries, and shares her own research questions, including why people rate AI-generated therapy and art highly, right up until they discover it was AI-generated. The episode closes with a discussion of AI regulation across the EU, China, the US, and the UK, and the unresolved question Kate most wants answered: who are the millions of people in relationships with AI companions, and what are they really getting out of it? Available on Apple Podcasts, Spotify, YouTube, and via RSS. https://www.kcl.ac.uk/people/kate-devlin [https://www.kcl.ac.uk/people/kate-devlin]

95% of AI Programs Fail. The Strategist Who Dares to Say Why | Dr. Anastassia Lauterbach

She arrived in Germany at 20 years old with 50 Deutsche Marks, no visa for the country she actually wanted to reach, and no safety net. Three decades later, Dr. Anastassia Lauterbach is a Global Hall of Fame honouree for business excellence, a former CEO of Qualcomm Europe, author of The AI Imperative, and one of the clearest, most unsparing voices in the global AI conversation. In this episode of Empathy Unbound, Andrew Phipps sits down with a strategist who will tell you what is actually happening in AI without flinching. Anastassia explains why 95% of AI programmes fail, why current LLMs cannot reason, why hallucinations are built into the architecture and cannot simply be engineered away, and why she rates the chance of AI consciousness at zero. She also makes the case that AI is not a technology problem but a leadership problem, and that the solution must come from the bottom up, not the top down. Anastassia is self-funding a global AI literacy study whose results will be presented at the House of Lords on 22 October. She shares what young people in that study are actually saying — including the striking finding that they feel "forced to learn AI from TikTok." She also tells the story of what she said to her daughter at four and a half years old about working hard to have a choice, a message that daughter is still repeating today. This is a conversation about real intelligence, human agency, and the courage required to say what others in the room are afraid to say. IN THIS EPISODE: • Why she left Moscow at 20 with 50 Deutsche Marks and how that shaped her ethical lens • The AI knowledge gap she spotted in US boardrooms in 2017 and why it has barely closed • Why 95% of AI programmes fail and why having tools is not a strategy • Why hallucinations cannot be solved, only reduced — and what the architecture makes inevitable • Why AGI will not come from the current LLM labs • Her verdict on AI consciousness: zero, and the neuroscience behind it • Why AI has become a new religion, complete with churches, apostates, and believers • The global AI literacy study she is self-funding, with results at the House of Lords in October • "We are forced to learn AI from TikTok": what young people in her study actually said • What she told her four-and-a-half-year-old daughter about working hard to have a choice • Why AI is not a technology issue, it is a leadership issue • Europe's demographic decline and why AI literacy is not optional • Her mission: one million families and one hundred thousand businesses ABOUT DR. ANASTASSIA LAUTERBACH: Global Hall of Fame honouree for business excellence. Former CEO of Qualcomm Europe. Author of The AI Imperative (2018). Founder of AI Edutainment and creator of the Romy and Roby universe. Host of the AI Snacks with Romy and Roby podcast. Mission: one million families and one hundred thousand businesses. Website: http://www.aiedutainment.ai/ [http://www.aiedutainment.ai/] ABOUT EMPATHY UNBOUND: Hosted by Andrew Phipps, exploring how empathy shapes leadership, technology, society, and the way we live. New episodes weekly on Apple Podcasts, Spotify, YouTube, and RSS. Search Empathy Unbound wherever you listen.

Shea Murtaugh on Leading Through Change

Shea Murtaugh on Building a Resilient Marketing Agency, Leading Through Change, and Using AI as a Strategic Advantage Shea Murtaugh, founder of a nationally recognized integrated marketing agency launched in 2004 after leaving broadcast sales, reflects on her blue-collar upbringing, early “C student” years, and the commission-sales mindset that gave her confidence to go all-in as an entrepreneur. She describes major setbacks—an employee taking clients, steep revenue drops, resigning a largest client during COVID—while emphasizing resilience as showing up “the next morning,” and credits long-term client retention to chemistry, trust, measurement, and preparing clients for change. Murtaugh explains her shift from doer to CEO by hiring experts, joining Women Presidents Organization, and building a business that can run without her, while sustaining a virtual, 40-hour work culture with strong perks. She sees AI as the biggest industry shift, boosting output, changing pricing models, and driving new initiatives like Leap Shift AI and an LLM-agnostic tool to automate reporting and free teams for strategy. 00:00 Meet Shea Murtaugh 00:43 Formative Years 02:44 Taking the Leap 06:19 Resilience Lessons 09:06 Keeping Clients Long Term 10:25 From Doer to CEO 13:07 Choosing the Right Clients 15:12 AI Changes Everything 17:32 Teaching AI Adoption 20:54 Building Ada Tool 24:19 Future of Media Marketing 29:59 Brand Niches and Giants 32:40 Ads Everywhere Now 33:51 Targeted Ads Tradeoff 35:42 Defining Real Success 37:52 Never Selling The Business 40:43 Rethinking Work Life Balance 43:14 Leadership Style Evolution 45:29 Life Experiences That Shaped Her 49:05 Equality Beyond ESG Talk 50:36 Self Care As A Founder 54:18 Scaling Up With AI 56:43 Advice For Young Entrepreneurs https://hoffmannmurtaugh.com/ [https://hoffmannmurtaugh.com/] https://www.linkedin.com/company/hoffmannmurtaugh/ [https://www.linkedin.com/company/hoffmannmurtaugh/] https://www.instagram.com/hoffmannmurtaugh/ [https://www.instagram.com/hoffmannmurtaugh/] https://www.tiktok.com/@hoffmannmurtaugh [https://www.tiktok.com/@hoffmannmurtaugh]? https://www.facebook.com/HoffmannMurtaugh/ [https://www.facebook.com/HoffmannMurtaugh/] https://twitter.com/HMurtaugh?lang=en [https://twitter.com/HMurtaugh?lang=en] https://www.pinterest.com/hoffmannmurtaugh [https://www.pinterest.com/hoffmannmurtaugh]

Metabolic Health in Midlife with Lorie Eber

Metabolic Health in Midlife: Stress, Sleep, Habits, and the Reality of Weight Loss Drugs with Lorie Eber Andrew interviews Lorie Eber, a former corporate litigator who retired at 49 and later became a board-certified health and wellness coach focused on women in midlife. Eber describes leaving an identity built on high-stress work and finding a more empathetic path in wellness. They discuss how modern food and sedentary environments, technology-driven stress, and disconnection undermine health, arguing that “discipline” issues are often exhaustion and context. Eber outlines metabolic health as an integrated brain-body system shaped by stress, sleep, social connection, food quality, hydration, and movement “exer-snacks,” noting menopause is under-taught in medical training. They address GLP-1 weight-loss drugs as helpful but risky without lifestyle change, and emphasize slow, specific goals, accountability, mindful eating, and controlling the home food environment amid perfectionism and unsupportive corporate cultures. 00:00 Welcome and Introduction 00:42 From Lawyer to Coach 02:17 Stress Identity and Change 08:46 Discipline vs Environment 10:31 Movement and Modern Life 14:36 Technology Stress and Loneliness 20:03 Midlife Menopause Gaps 22:05 Metabolic Health Explained 27:57 GLP-1 Drugs Pros and Cons 33:42 Social Life on GLP-1s 34:08 Side Effects and Who It Helps 34:36 PCOS Hype and Pharma Greed 35:50 Unknown Long Term Risks 36:52 Breaking Perfectionism Patterns 40:23 Corporate Culture vs Wellness 44:54 Retirement and Lost Community 46:16 Goals and Accountability 48:49 Control Your Food Environment 58:12 Mindful Eating and Satiety 01:02:21 Purpose and Closing Thoughts https://lorieeberwellnesscoaching.com/ [https://lorieeberwellnesscoaching.com/] https://www.facebook.com/LorieEberWellnessCoaching [https://www.facebook.com/LorieEberWellnessCoaching] https://twitter.com/EberLorie/ [https://twitter.com/EberLorie/] https://www.youtube.com/channel/UC7CeLbLNKnAva9szHp3Jx3A [https://www.youtube.com/channel/UC7CeLbLNKnAva9szHp3Jx3A] https://www.linkedin.com/in/lorieeber [https://www.linkedin.com/in/lorieeber] https://www.tiktok.com/@lorieeber/ [https://www.tiktok.com/@lorieeber/] https://www.instagram.com/lorieeber/ [https://www.instagram.com/lorieeber/]

Ethical, Human-Centered AI with Robert Vandervoort

Ethical, Human-Centered AI: Accessibility, Cost, and Embodied Agents with Robert Vandervoort Robert Vandervoort, founder of VDV Labs, shares how early curiosity with technology, ADHD, legal troubles, and living with optic nerve atrophy and legal blindness shaped his empathy-driven approach to ethical, human-centered AI. He describes moving from help desk and operations into solutions engineering, and how employee-experience work at Moveworks influenced his thinking about AI as augmentation rather than replacement. The conversation explores why many AI initiatives fail without clear business problems, how companies chase hype, and where AI can add value in IT, customer service, manufacturing, and healthcare (e.g., triage, agent assist, transcription, computer vision, radiology support). They discuss AI’s real costs (tokens, hardware, memory, water and heat impacts), trust in self-driving, AI companions and grief bots, and Vandervoort’s “Chip” project toward embodied, agentic AI plus other VDV Labs products. 00:00 Meet Robert Vandervoort 00:55 Curiosity and Early Tech 03:09 ADHD and School Struggles 05:56 Rock Bottom and Blindness 08:10 Back to Computers 11:53 Discovering AI in 2022 12:28 Solutions Engineer Path 16:57 Human Centered Ethics 20:19 Selling With Empathy 23:03 Efficiency vs Opportunity AI 24:24 Cost Cutting and Consequences 32:10 AI Costs and Tokens 33:43 Value vs Expensive 34:11 Memory Prices and AI Demand 35:51 Water Cooling and Hidden Costs 38:35 Space Data Centers Debate 40:23 Why AI Projects Fail 48:21 Where AI Fits in Business 48:58 Practical AI Use Cases 57:45 AI Companions and Consciousness 58:32 Embodied Agents and Senses 01:00:45 Biology vs Neural Nets 01:03:00 Memory Tricks and Proprioception 01:04:13 Consciousness And Context 01:05:18 CHIP Vision System 01:07:41 Introspection And Memory 01:09:28 Embodiment And Agency 01:14:57 Commercializing The Robot 01:17:02 Companion AI Risks 01:21:40 Grief Bots Debate 01:26:54 VDV Labs Next Steps 01:30:21 Closing Thoughts https://www.linkedin.com/in/rvandervoort/ [https://www.linkedin.com/in/rvandervoort/] https://vdvlabs.ai/ [https://vdvlabs.ai/]