The AI Values Podcast

Ep. 7 | Agentic AI Rewrites Your Whole Business | The AI Values Podcast

What if 40% of agentic AI projects are already heading for failure and the first warning sign isn't an error. Instead, it is tension? Edosa Odaro [https://www.linkedin.com/in/edosa/] and Lindley Gooden [https://www.linkedin.com/in/lindleygooden/] go head-to-head on Episode 7 of The AI Values Podcast for the most uncomfortable conversation we've had yet on agentic AI in the enterprise the over-optimisation trap, the human capacity gap, and the one analogy every C-suite needs before bringing AI agents into the business. WHAT IS COVERED: ► Why agentic AI is like bringing George Clooney into a movie the new-actor analogy that explains why AI agents change the whole system, not just a workflow ► Why the first thing agentic AI delivers isn't value it is tension, friction, and the human cost of misaligned speed ► Agentic AI over-optimisation: the 2:30am warehouse moment when the AI didn't fail it worked too well ► Misalignment vs malfunction: the governance distinction every C-suite is missing about agentic AI risks ► Why agentic AI doesn't work in isolation it works inside an ecosystem of humans, processes, and pace ► The trade-off problem: what your AI is optimising against, and the value quietly leaving the building ► AI value vs values: the one question every board should ask before signing the next agentic AI budget line ► Why "trust forward, verify always" is the only credible posture for AI agents in business ► AI agents and workforce transformation what the "we'll just retrain them into data scientists" story really hides ► A practical C-suite guide to preparing your business for agentic AI before the friction shows up ⏰ EPISODE TIMESTAMPS: 0:00 [https://www.youtube.com/watch?v=a_uTBRoGdFI] — Cold Open: Agentic AI Is the New Actor in Your Movie 0:35 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=35s] — Bringing AI Agents Into Your Business: Why a Tool Mindset Breaks 2:09 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=129s] — If It's Transformative, How? Why Tension Shows Up Before Value 4:35 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=275s] — The 2:30am Warehouse: When the Agent Runs Faster Than Its Humans 6:40 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=400s] — Trust, Verify, Communicate: The Honest Conversation Agentic AI Demands 9:04 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=544s] — Tension and Trade-Offs: The Two Things Every Agentic System Brings 11:10 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=670s] — Value vs Values: The Real Question Every Board Should Be Asking 12:20 [https://www.youtube.com/watch?v=a_uTBRoGdFI&t=740s] — Stay in the Conversation: How to Get Involved

Ep. 6 | Why AI Projects Really Fail | Graeme McDermott | The AI Values Podcast

What if 95% of AI projects aren't really failing and the actual problem is that no one defined what success looked like in the first place? Graeme McDermott [https://www.linkedin.com/in/chiefdataanalyticsofficerlondon/], Chief Data Officer at Tempcover, with two decades leading data and analytics functions across the AA, Addison Lee and Tempcover joins Edosa Odaro and Lindley Gooden on Episode 6 of The AI Values Podcast for one of our most pragmatic boardroom-level conversations to date. On AI accountability, the rise of "AI told me" decision-making, and what really happens to graduate jobs when companies cut their next generation of leaders to fund the LLM bill. Graeme unpacks why every C-suite is quoting the same 85–95% AI failure stat and why most of that "failure" is actually a definition-of-success and data-foundations problem, not a technology problem. Bad questions have no context. Lazy AI use, trains lazy thinking. And cutting the entry-level rung doesn't save money it deletes your future bench. WHAT WE COVER: ► Why 95% of AI projects "fail" and the leadership reframe (you didn't define success) ► "This is correct because ChatGPT said so" when AI becomes the decision ► Bad questions have no context the prompting skill C-suites are skipping ►AI accountability in the boardroom: who is for it, who is resisting, who is responsible ►Cognitive offloading the MIT research on why AI users disengage their brains ►Entry-level collapse: the 25% drop in UK graduate jobs and the bench-strength crisis ►Apprenticeships, electricians, and rebuilding the skills ladder for the AI era ►What every board should take away, one thing the C-suite must own ⏰ EPISODE TIMESTAMPS:  00:00 — Cold open & welcome 06:40 — Excited or concerned? An insider's view of AI today (Graeme McDermott) 10:49 — "AI told me" when ChatGPT becomes the cited authority 13:54 — Trust, prompts, and the AI-lazy five-word problem 17:00 — Why 85–95% of AI projects "fail" and what success actually means 22:40 — The boardroom: who's for AI, who's resisting, who's accountable 27:45 — One thing every board should take away learn a trade 30:15 — Debrief & sign-off ABOUT THE AI VALUES PODCAST: The AI Values Podcast is where leaders come to think clearly about the trade-offs behind AI adoption not just the opportunities. Hosted by Edosa Odaro (author, 'The Values of AI') and Lindley Gooden (author, 'The Future of Truth'), with weekly conversations at the intersection of AI, trust, governance, and the future of work. 🎙 SUBSCRIBE to The AI Values Podcast for honest, rigorous conversations at the intersection of AI ethics, AI governance, and business leadership. ◼ Find out more: https://www.theaivalues.org [https://www.theaivalues.org] ◼ Reach out: podcast@theaivalues.org [podcast@theaivalues.org] ◼ Get the Weekly AI Values Dispatch → https://pages.theaivalues.org [https://pages.theaivalues.org] ◼ Edosa Odaro: https://www.linkedin.com/in/edosa/ [https://www.linkedin.com/in/edosa/] ◼ Lindley Gooden: https://www.linkedin.com/in/lindleygooden/ [https://www.linkedin.com/in/lindleygooden/] ◼ Guest: Graeme McDermott: https://www.linkedin.com/in/chiefdataanalyticsofficerlondon/ [https://www.linkedin.com/in/chiefdataanalyticsofficerlondon/]

Ep. 5 | Value Erosion. The AI Failure Boards Never See | The AI Values Podcast

What if 68% of AI projects have already failed and the boardroom is the last to know? Edosa Odaro [https://www.linkedin.com/in/edosa/] and Lindley Gooden [https://www.linkedin.com/in/lindleygooden/] go head-to-head on Episode 5 of The AI Values Podcast for the most uncomfortable conversation we've had on AI investment yet, especially on value erosion, the capability paradox, and the one question every board should be asking before signing the next AI budget line. 🎯 Three questions to bring into your next board meeting drawn straight from this conversation. ────────────────────────────────────── THIS WEEK: A HEAD-TO-HEAD No guest. Just Edosa and Lindley working a single problem from two angles — Edosa with the practitioner's frame, Lindley with the boardroom question that won't go away. The recent RAND statistic on the table; the argument working out in real time. ────────────────────────────────────── WHAT WE COVER:  ► Why 68% of AI projects quietly stop delivering value within six months and look like success while they do  ► Value erosion: the term Edosa coins for what the boardroom never sees  ► The capability paradox: why your organisation can't absorb the AI capability you've invested in  ► How 90% predictive accuracy can deliver zero value the metrics trap that catches CFOs every time  ► When bad data becomes wrongful detention the real cost of unaccountable AI  ► Why "stop AI" isn't a credible governance strategy and what meaningful friction looks like instead  ► The one question every AI investment should answer: who benefits, who loses? ⏰ EPISODE TIMESTAMPS:  0:00 — Introducing Episode 5: The 68% Stat That Should Worry Every Board  2:21 — The Capability Paradox: Why Organisations Can't Absorb AI  4:09 — 90% Accuracy, Zero Value: The Metrics Trap  5:57 — The One Question AI Should Answer: Are Decisions Improving?  7:16 — When Bad Data Becomes Wrongful Detention  9:42 — Why "Break Down All the Silos" Is the Wrong Answer  11:10 — Should We Stop AI? Edosa Pushes Back  14:16 — Guardrails That Actually Work — Not Brakes  15:09 — Who Benefits, Who Loses: The Question Every Board Should Be Asking  19:04 — Communication Is the Unlock #AIgovernance #CapabilityParadox #ResponsibleAI #ValueErosion #BoardroomAI #AIROI #AIaccountability #EthicalLeadership #AIRiskManagement #AIstrategy

Ep. 4 | More People Will Switch Off From AI | The AI Values Podcast ft. Dr Naeema Pasha

What if the biggest AI risk isn't bias or hallucination but the quiet moment where people simply switch off? Dr Naeema Pasha joins Edosa Odaro and Lindley Gooden on Episode 4 of The AI Values Podcast for one of our most human conversations yet on trust, fairness, AI literacy, and the leadership shift that will decide who wins the next decade. 🎯 Three test-yourself moments in this episode, see how well you read the research before Dr Pasha reveals the answers. ────────────────────────────────────── GUEST: Dr Naeema Pasha [https://www.linkedin.com/in/dr-naeema-pasha-9b23b66/] Dr Pasha is a Henley Business School academic, author, and one of the UK's leading voices on Industry 5.0 and human-centred AI. She unpacks why we still trust humans over AI, why confident people use AI better, and why the Henley 'Equity Effect' research found a 58% revenue premium for organisations that actively pursue fairness. WHAT WE COVER: ► Why this AI moment matters and the 'They Not Like Us' problem Fairness as the driver word for AI governance ► Why trust in AI is already eroding and what leaders miss AI coaching, therapy bots, and the 60% of UK adults already using them ► The Henley Equity Effect: a 58% revenue premium for fairness-led organisations ► Industry 5.0, the cult of the CEO, and empathy-based leadership ► AI literacy, critical thinking, and why confidence beats mastery ► The quiet AI backlash 'no AI used here' signs, and what they mean for brand trust ABOUT THE AI VALUES PODCAST: The AI Values Podcast is where leaders come to think clearly about the trade-offs behind AI adoption not just the opportunities. Hosted by Edosa Odaro (author, 'The Values of AI') and Lindley Gooden (author, 'The Future of Truth'), with weekly conversations at the intersection of AI, trust, governance, and the future of work. 🎙 SUBSCRIBE to The AI Values Podcast for honest, rigorous conversations at the intersection of AI ethics, AI governance, and business leadership. ◼ Find out more: https://www.theaivalues.org [https://www.theaivalues.org] ◼ Reach out: podcast@theaivalues.org [podcast@theaivalues.org] ◼ Get the Weekly AI Values Dispatch → https://pages.theaivalues.org [https://pages.theaivalues.org] ◼ Edosa Odaro:   / edosa   [https://www.youtube.com/redirect?event=video_description&redir_token=QUFFLUhqbm15NVg3X1ZZTVQ2UU5GczJjT1VYaVdqSlhoUXxBQ3Jtc0tsZDlzM2ZnXy1UTmJqYU1NU0lEWGJtZUV5aUNEbFE1ZUV6SFJVdEJjRDUyX2J2c0FPd3U5Z3Z2Y1lwMkpVeU1obllhbGNwZ3d2UkVuVUZuMVRSeVR5eVpBejY1a01rUUlhSThndldhcEVzbDVub3dCQQ&q=https%3A%2F%2Fwww.linkedin.com%2Fin%2Fedosa%2F&v=vBSK1Il_SaU] ◼ Lindley Gooden:   / lindleygooden   [https://www.youtube.com/redirect?event=video_description&redir_token=QUFFLUhqbWp0TkhFX3VRclVJbHp2VFRVYkJEbEFXbUhrZ3xBQ3Jtc0tsc3dSSmJNeG1udDFwcXQ1UTRDZWRkVlRCZzFpMi1SZG95bkJMOWRkQU13UzJKVDZuQkgwYjJlTjFsQWNWYldmbWlMbGU2MWEyRnRZRXpTQVIzTnRnRzd4ME9USkRvN2l3ZDFFdFVTU1hETExIXzJLbw&q=https%3A%2F%2Fwww.linkedin.com%2Fin%2Flindleygooden%2F&v=vBSK1Il_SaU]

Ep. 3 | Are We Trusting AI Too Much? | The AI Values Podcast

Confident doesn't mean correct. And the space between those two things can have real, serious consequences. In this episode of the AI Values Podcast, Edosa Odaro [https://www.linkedin.com/in/edosa/]and Lindley Gooden [https://www.linkedin.com/in/lindleygooden/] ask one of the most important questions of the AI age: are we trusting AI too much? Through two real and gripping stories a family member who ended up in hospital after an AI gave dangerous food safety advice for a child with a nut allergy, and a journalist who nearly published a fabricated quote generated by an AI chatbot this conversation challenges the assumptions most of us carry every single day. You'll hear about: • Why AI sounds authoritative even when it's completely wrong • How AI hallucinations continue to pose serious, real-world risks • Why executives and business leaders are especially vulnerable to over-trusting AI • How to build genuine critical thinking skills in an AI-first world • What we risk losing cognitively and professionally when we stop questioning AI Whether you use AI at work or at home, this episode will make you think before you trust. About The AI Values Podcast: Where senior leaders come to think clearly about AI not just what it can do, but what it should do, and for whom. Because AI can deliver value without losing what we value most. More at theaivalues.org [http://theaivalues.org]