Ep. 31 | Why Give AI Agents Autonomy We Don't Give Employees? (Guest: Lasse Rindom)

Why are companies so eager to give AI agents autonomy when their human employees operate within carefully defined rules? Lasse Rindom has a question for the agentic AI movement, and it goes straight to how enterprises actually work. Organizations depend on structure, accountability, and coordination. Giving an AI agent unlimited freedom to improvise might sound impressive, but what happens when that agent has to work alongside people, processes, and systems that cannot change overnight? In a wide-ranging and often hilarious conversation, Ian Barkin and Lasse explore the laws of enterprise physics, why handing every employee an AI license does not automatically transform a business, and what Monty Python's machine that goes "ping" can teach us about technology investments. They also challenge the idea that friction is always something to eliminate, examine whether regulation can actually help innovation move faster, and make the case for keeping human judgment, critical thinking, and genuine interaction at the center of AI adoption. **About the Host** Ian Barkin is the co-founder and co-CEO of magentIQ, with decades of experience in intelligent automation, outsourcing, and operational transformation. Through The magentIQ Show, he brings together business leaders, technologists, and industry thinkers to explore the practical realities of AI adoption and the future of work. **About the Guest** Lasse Rindom is the Lead AI & Technology Strategist at Basico and host of *The Only Constant*, a podcast exploring AI, business, and enterprise technology adoption. With a background in history, robotic process automation, and enterprise transformation, he brings an unusually philosophical yet practical perspective to the AI conversation. His work focuses on connecting emerging technologies with real business problems, organizational change, and human decision-making. **Explore More from Lasse Rindom** **The Only Constant Podcast** Conversations with technology and business leaders on AI implementation, organizational change, and the realities of enterprise transformation. https://theonlyconstant.podbean.com/ **Lasse Rindom on LinkedIn** https://www.linkedin.com/in/lasserindom/ **Lasse's Article: From Broad Chat to Specialised and Useful AI Function** https://basico.dk/en/inspiration/blog/from-broad-chat-to-specialised-and-useful-ai-function **Books and References Mentioned** **The Civilizing Process by Norbert Elias** Lasse draws on Elias's work to explain how growing interdependence shapes behavior, coordination, and the constraints that allow large organizations to function. https://www.wiley-vch.de/de?isbn=9780631221616&option=com\_eshop&title=The+Civilizing+Process&view=product **The Protestant Ethic and the Spirit of Capitalism by Max Weber** Referenced during Ian and Lasse's discussion of cultural differences, work, and attitudes toward rules and responsibility. https://www.routledge.com/The-Protestant-Ethic-and-the-Spirit-of-Capitalism/Weber/p/book/9780415254069 I**n This Episode** (1:20) Two old friends reunite, and the story of Ian becoming Lasse's very first podcast guest (2:07) The laws of enterprise physics: why organizations can only change so fast (4:05) From history and poetry to RPA, AI strategy, and 100 episodes of The Only Constant (7:00) The AI rainforest: building autonomous agents without understanding what they might do (8:01) Novelty fatigue, tech deflation, and the challenge of keeping up with new AI models (11:36) The fundamental question every AI project should answer: what problem are we solving? (14:21) Monty Python and the machine that goes "ping": when the technology becomes the attraction (15:12) Visible AI versus useful AI, and why the best solution may have no chatbot at all (18:30) Revisiting old automation opportunities that new AI capabilities could finally make possible (22:00) Knowledge attrition, forgotten decisions, and building better organizational memory (24:04) Why giving every employee an AI license will not transform an enterprise (26:06) Forward-deployed engineers and the challenge of finding people who understand business and technology (28:15) Better AI models demand better questions, and the importance of critical thinking (29:09) Why give AI agents autonomy that enterprises would never give human employees? (31:05) Deterministic work, governance, and why stability matters in enterprise operations (34:11) The global AI race, European adoption, and where the real value of AI models might emerge (36:23) Can regulation accelerate innovation? Lasse makes the case using Denmark and employment contracts (42:03) Why bridging business and IT requires understanding culture, accountability, and competing priorities (45:24) Norbert Elias, the civilizing process, and the organizational constraints that make enterprises work (47:57) The case for good friction: why removing every obstacle may remove something valuable (49:00) Cognitive offloading, learning through struggle, and why AI-generated work does not always stick (50:23) Helping people with AI behind the scenes instead of forcing them to interact with chatbots (55:16) Practical AI use cases: preserving institutional knowledge and improving finance and HR workflows (56:16) Using AI to improve answers while keeping company policies accurate and up to date (58:00) The value of genuine human conversation in an increasingly AI-generated world (1:03:36) Robert Frost, the road not taken, and finding perspective amid technology FOMO Listen now, and tell us: how much autonomy would you actually give an AI agent in your organization? Find Us on Your Favorite Streaming Service iTunes: https://podcasts.apple.com/us/podcast/the-magentiq-show/id1896570951 Spotify: https://open.spotify.com/show/033f2oKxnFr5fmcpdSHjSL iHeartRadio: https://iheart.com/podcast/333292440 Got a perspective worth sharing? We are always looking for guests. Reach out at [info@bemagentiq.com](mailto:info@bemagentiq.com).