Beyond The Prompt - How to use AI in your company

Hosted by Jeremy Utley & Henrik Werdelin

Business Entrepreneurship EducationInterviews guests

Website RSS feed

Episodes

Latest episode

Aug 2026

Language

About the show

Beyond the Prompt dives deep into the world of AI and its expanding impact on business and daily work. Hosted by Jeremy Utley of Stanford's d.school, alongside Henrik Werdelin, an entrepreneur known for starting BarkBox, prehype and other startups, each episode features conversations with innovators and leaders to uncover pragmatic stories of how organizations leverage AI to accelerate success. Learn creative strategies and actionable tactics you can apply right away as AI capabilities advance exponentially.

Listen to episodes

60 recent

August 19, 20261 hr 5 min

What Happens When AI Adoption Actually Works? - with Eric Porres, Chief AI Officer of Logitech

A year ago, Logitech had people experimenting with AI. Today, Eric says he can’t think of a single part of the company that isn’t building, exploring, or creating something with it. Eric shares what helped make that happen. There’s a Build Advisor that helps employees figure out what to build and connects them with people who may have already worked on something similar. AI in Action moments are now part of company and leadership meetings. And before leadership presents to the board, there’s an expectation that their work goes through an AI Board Advisor first. Henrik, Jeremy, and Eric also get into what comes next: how to measure whether all this AI activity actually creates value, why Eric built AI systems to manage his own information overload and sleep, and why creating something new should come with another question: what old report, process, or way of working can now disappear? Key Takeaways Embed AI into how work gets done AI becomes more valuable when it’s built into existing workflows and expectations, rather than simply made available for people to use. Make AI adoption visible and repeatable Logitech keeps AI present through AI in Action moments, leadership routines, office hours, shared Gems, and a 175-person volunteer Champions Network. Build resources that help people help themselves Tools like the Build Advisor give employees a place to start, surface work that already exists, and connect them with colleagues who have tackled similar problems. Ask what you can stop doing Eric argues that every new AI-enabled artifact should come with another question: what old report, process, or way of working can now disappear? Eric's website: porres.com/ Eric's LinkedIn: linkedin.com/eporres/ Logitech: www.logitech.com/ 00:00 Embedding AI Into the Workflow 00:52 Meet Eric Porres 01:15 The Cambrian Explosion of AI 06:20 Measuring the Value of AI 11:16 The Build Advisor 16:12 Keeping Up With AI 19:43 Making AI Part of the Culture 21:37 The AI Board Advisor 25:51 Building an AI Champions Network 29:15 Eric’s Personal AI Stack 32:17 The AI Vampire Problem 40:59 Building a Deep Memory 46:49 What Can AI Help You Delete? 55:19 The Debrief 📜 Read the transcript for this episode: Here! For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

August 5, 202643 min

Why Companies That Are Great at Innovation Still Fail - with Stanford Professor Charles O'Reilly

Most companies know how to innovate. Far fewer know how to scale innovation. Charles introduces the explore versus exploit framework, explaining why the same systems that help organizations succeed today can make them resistant to change tomorrow. As companies mature, they become better at serving existing customers, improving existing products, and optimizing existing processes. The harder question is how to create space for experimentation without undermining the business that already exists. Henrik, Jeremy, and Charles explore what this means in the age of AI. They discuss whether AI should be viewed as a substitute or a complement, why Microsoft's transformation under Satya Nadella succeeded, how Amazon has built exploration into its operating system, and why leaders don't create adaptable organizations through vision alone. They do it by shaping culture through incentives, systems, and the behaviors they reward. Key Takeaways: The biggest challenge isn't generating ideas. It's scaling them. Many organizations are good at innovation. The difficult part is giving promising ideas the support they need to grow. Great companies become trapped by what made them successful. The metrics, incentives, and culture that optimize today's business can make it harder to adapt to tomorrow's. Culture is built through systems. Leadership principles only matter when they're reflected in hiring, incentives, performance reviews, and everyday behavior. The goal isn't to predict the future. It's to discover it. The most adaptable organizations build processes that help them experiment, learn, and uncover new opportunities as the world changes. Charles' Stanford profile: stanford.edu/faculty/charles-oreilly 00:00 Intro: Why Companies Die Fast 00:36 Meet Charles O'Reilly 02:01 Explore Versus Exploit 03:50 AI Substitute Or Complement 06:51 Adaptability As Culture 09:19 Resistance To Change 12:32 Microsoft Culture Turnaround 15:25 Ambidexterity And Lifespans 18:20 Ideate Incubate Scale 20:10 Scaling Needs Separation 24:44 Amazon PRFAQ Machine 34:33 Rituals And Failure Signals 38:05 The Debrief 📜 Read the transcript for this episode: here! For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

July 22, 202650 min

What You Can Learn From Asking AI for Brutal Honesty About Yourself - with Bestselling Author Daniel Pink

The conversation begins with an experiment that caught Jeremy's attention. Dan asked ChatGPT to tell him what his friends wouldn't. From questions about his blind spots to what people might say behind his back, some responses felt completely wrong, while others landed with surprising force. Rather than accepting every answer, Dan explains why the real value comes from wrestling with AI's perspective, not simply believing it. From there, Henrik, Jeremy, and Dan explore what it takes to use AI well. They discuss intellectual humility, prompting models to challenge rather than flatter us, and why AI works best as a sparring partner that exposes weaknesses in our thinking. The conversation also touches on taste, agency, and why, in a world where execution is becoming easier, discernment and original thinking become even more valuable. Key Takeaways: Use AI to make your thinking visible The best AI conversations don't just generate answers. They help you understand your own assumptions, reactions, and ideas more clearly. Treat AI as a sparring partner Challenge AI's responses, ask it to critique your work, and use it to strengthen your thinking rather than replace it. Taste is developed by creating As AI makes execution easier, judgment and discernment become more valuable. The best way to develop both is by creating, not just consuming. Agency depends on context People don't become more agentic through willpower alone. The right environment, with autonomy and room to take risks, makes initiative possible. Daniel's website: danpink.com LinkedIn: linkedin.com/danielpink AI Self-Reflection Prompts: danpink.com/ai-guide/ 00:00 AI as a Brutally Honest Advisor 00:32 Meet Daniel Pink 00:47 AI for Self-Knowledge 01:40 A Brutally Honest AI 06:10 What Do People Say Behind Your Back? 07:45 Should You Trust AI's Advice? 13:28 The Fear of Irrelevance 17:44 Intellectual Humility 20:04 AI as a Sparring Partner 24:35 Teaching AI to Think Like You 27:38 Why Taste Matters 29:59 Agency Starts with Context 37:08 A Future That's a Little Better 41:49 Nostalgia vs. Reality 45:18 The Debrief 📜 Read the transcript for this episode: Here! For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

July 8, 202653 min

Are You Qualified to Challenge Your Team on AI? - with Geoff Woods, Author of The AI-Driven Leader

Geoff Woods returns to Beyond the Prompt to discuss the updated edition of The AI-Driven Leader and what has changed over the past 18 months. Rather than focusing on the latest AI models, Geoff argues that leaders need to use AI themselves before asking others to, using it to think more clearly, shape strategy, and make better decisions. The conversation explores why many organizations confuse access with adoption, why strategy should come before use cases, and how AI can change the way leaders approach everything from business strategy to organizational design. Along the way, Henrik and Jeremy challenge Geoff's ideas on authorship, judgment, and whether understanding AI changes what leaders believe is possible. Key Takeaways: Leaders need to use AI themselves Using AI personally is what qualifies leaders to shape strategy and lead others from practice rather than theory. Use AI to improve your thinking The biggest opportunity isn't automating work. It's using AI to think better, solve better problems, and imagine new possibilities. Start with problems, not use cases Begin with the biggest challenges facing the business, then use AI to rethink how to solve them. AI still needs human judgment AI can generate ideas, but people are still responsible for reviewing the output and standing behind it. Focus AI on your highest-value work Use AI to amplify the small set of activities where your human strengths create the greatest impact. The AI-Driven Leader: aileadership.com Geoff's LinkedIn: linkedin/geoff-woods 00:00 Are You Qualified to Lead on AI? 00:35 Meet Geoff Woods 00:54 The AI Slop Dilemma 05:48 Putting Your Stamp of Approval 09:19 What Changed in 18 Months 12:13 Access Isn't Adoption 15:50 Why Leaders Can't Delegate AI 20:09 Strategy Before Use Cases 22:02 BarkBox's AI Strategy 26:23 Reinventing Strategy with AI 33:11 Compressing Months into Hours 37:22 Human Skills as Superpowers 42:46 The Debrief 📜 Read the transcript for this episode: are-you-qualified-to-challenge-your-team-on-ai-with-geoff-woods-author-of-the-ai-driven-leader/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

June 24, 202653 min

The Unexpected Economics of AGI - with Christian Catalini, Tech Founder and Co-Creator of Libra

Christian believes the AI era will be defined less by generating outputs and more by evaluating them. As intelligence becomes cheaper and more accessible, the people who create the most value may be those who can distinguish good work from exceptional work and help guide increasingly capable systems. The conversation explores verification, judgment, and why expertise still matters in a world where AI can perform many tasks at a high level. Christian explains why today's experts are both highly valuable and simultaneously training the systems that may eventually replace parts of their work. Jeremy and Henrik also explore what this means at a personal level. They discuss building AI agents that reflect your own preferences, creating personal verification systems, and why AI may make it easier to learn new skills, switch careers, and pursue more ambitious ideas. Key Takeaways: Verification becomes more valuable as intelligence gets cheaper As AI makes generating outputs easier, the ability to recognize what is actually good becomes increasingly important. Experts are training their own replacements The people best positioned to verify AI outputs are also helping codify the expertise that trains future systems. Human value shifts from doing to directing As AI handles more execution, people create value through judgment, direction, and orchestration. Build your own verification system The best AI users are developing agents, workflows, and tools that reflect their own preferences and standards. Christian's LinkedIn: linkedin.com/ccatalini/ Christian's X: x.com/ccatalini Christian's website: catalini.com Economics of AGI: full paper Jeremy's Persona File Template: YouTube/The8Files 00:00 Non-Measurable Frontiers 00:32 Meet Christian Catalini 01:08 The Economics of AGI 03:09 Why Verification Matters 06:37 Can Everything Be Measured? 10:32 The Rise of the Verifier 14:35 When Intelligence Gets Cheap 21:46 Building Your Verification Harness 24:08 Human + AI Augmentation 30:18 Persona Files and Privacy 33:12 Reasons for Optimism 36:00 Career Switching in the AI Era 39:31 The Debrief 📜 Read the transcript for this episode: the-unexpected-economics-of-agi-with-christian-catalini-tech-founder-and-co-creator-of-libra/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

June 10, 20261 hr 1 min

Why Fear Kills Curiosity and What That Means for AI - with Chantel Prat, Cognitive Neuroscientist

Chantel Prat studies how different brains make sense of the world. Her work starts from a simple idea: every experience leaves a mark. The inputs we consume shape how we think, what we notice, and ultimately who we become. The conversation explores why people often choose familiar rewards over uncertain opportunities to learn. Chantel explains the tension between exploration and exploitation, why curiosity is essential for growth, and how fear can prevent us from engaging with new technologies like AI. They also discuss theory of mind, cognitive offloading, and what happens when we increasingly rely on AI for thinking. The goal is not simply to do better work, but to use AI in ways that help us become better versions of ourselves. Key Takeaways: Curiosity requires safety When people feel threatened, they become defensive rather than exploratory. Fear gets in the way of learning. Better inputs create better outputs Every experience leaves a footprint on the brain. The ideas, conversations, and information we consume shape how we think and who we become. We naturally favor certainty over exploration Our brains are biased toward familiar rewards, even when something new may offer greater long-term value. Curiosity starts with admitting you might be wrong Learning requires recognizing that you do not already have the answer. Without that openness, exploration never begins. Use AI to become better, not just produce more The most important question is not what AI can do for you, but what you still want to get better at yourself. Chantel Prat: linktr.ee/chantelprat The Neuroscience of You: The-Neuroscience-of-you/book 00:00 Curiosity Versus Threat 00:31 Meet Chantel Prat 01:02 Why Input Shapes Brains 04:08 The Output Pressure Trap 05:52 Exploration Versus Exploitation 10:05 Average Brains And Teams 15:35 Theory Of Mind Defined 22:12 Practicing With AI Feedback 24:31 Offloading Thinking To AI 29:50 Humans In The Loop 35:16 Age And Tech Reactions 42:15 Why Curiosity Requires Safety 48:15 Personal Codex And AI 50:54 Becoming More Yourself 54:34 The Debrief 📜 Read the transcript for this episode: why-fear-kills-curiosity-and-what-that-means-for-ai-with-chantel-prat-cognitive-neuroscientist/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

May 27, 202656 min

Why Your Favorite Brand Stopped Caring About You - Eric Ries, Author of The Lean Startup

Eric Ries, author of The Lean Startup and the newly released Incorruptible , joins Beyond the Prompt to explore why most companies drift from their original mission over time. The conversation dives into governance, shareholder primacy, Anthropic’s unusual structure, and why AI makes these questions more important than ever. Eric Ries argues that most companies are built on a contradiction. Founders say they care about customers and impact, but legally, the company is structured to serve shareholders first. Over time, that mismatch tends to win. The conversation explores what that looks like in practice, why it is so hard to fix, and how a small number of companies have tried to design around it from the beginning. Eric reflects on advising Anthropic in its earliest days and what it actually takes to protect a mission as a company scales. A big part of the discussion is how governance gets treated as a legal formality when it is really a design problem. In the age of AI, Eric argues that the principles baked into a company’s structure early on may determine whether it stays true to its mission or slowly drifts away from it. Key Takeaways: Mission drift is often built in from day one Founders may say they care about customers and impact, but legally the company is structured to serve shareholders first. Over time, that mismatch tends to win. Governance is one of the highest leverage founder decisions If the structure is misaligned early on, founders can lose control of the company and its mission no matter how strong the original vision was. The system is stacked against mission-driven founders Even well-intentioned founders operate inside structures designed to prioritize short-term shareholder returns. Most do not realize it until it is too late. “Why not try?” is more powerful than it sounds Eric’s argument is not that fixing governance is easy. It is that most founders never even ask the question. AI makes this more urgent than ever As AI systems act more autonomously, the principles built into a company early on will shape whether it stays true to its mission or drifts away from it. Eric's new book: Amazon: Incorruptible Website: incorruptible.co Socials: X: x.com/ericries LinkedIn: linkedin.com/ericries The Lean Startup: theleanstartup.com 00:00 Mission vs Shareholder Value 00:32 Meet Eric Ries 01:37 Why Anthropic Needed Governance 06:47 The Long-Term Benefit Trust 10:00 Why Great Companies Drift 13:47 From Lean Startup to Incorruptible 18:14 Is It Too Late To Fix? 23:06 Governance As A Superpower 25:48 The Lies Founders Tell Themselves 28:49 The Rise Of Shareholder Primacy 33:09 The Unaccountability Machine 35:51 Profit vs Human Flourishing 37:24 The ROI Trap 38:26 The H-E-B Loyalty Story 41:14 Principles Beyond Metrics 42:54 AI, Thick Data, And Human Judgment 46:43 The Debrief 📜 Read the transcript for this episode: why-your-favorite-brand-stopped-caring-about-you-eric-ries-author-of-the-lean-startup/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

May 13, 202654 min

You Can't Outsource Wisdom: Bestselling Author Ryan Holiday on What the Stoics Have to Say About AI

Ryan Holiday argues that while AI can generate outputs, it cannot generate wisdom. Drawing on a story from Seneca about a Roman who used educated slaves to sound intelligent, he compares outsourcing thinking to outsourcing exercise: the value comes from becoming the kind of person who can do the work, not simply producing the answer. The conversation explores the difference between useful cognitive offloading and surrendering judgment entirely. Ryan explains that while tools like GPS may replace navigation skills without much consequence, writing, decision-making, and critical thinking shape the person on the other side of the process. AI, he argues, tends to amplify existing tendencies. People satisfied with mediocre work will settle faster, while people pushing for exceptional work can use AI to refine and challenge their thinking. Throughout the episode, Stoicism serves as a counterweight to both panic and hype. Change and uncertainty are constants throughout history, not exceptions. Ryan reflects on leadership, family, adaptability, and skepticism, arguing that in a world where AI can confidently produce both insight and nonsense, the ability to question, verify, and think independently becomes increasingly valuable. Key Takeaways: You cannot outsource wisdom AI can generate answers, but judgment and understanding still come from doing the work yourself. AI amplifies who you already are People who settle for mediocre work will do so faster with AI. People who push for better work can use it to deepen and refine their thinking. Bullshit detection is becoming a core skill As AI produces increasingly convincing answers, skepticism and verification become essential. Change is not new The Stoics viewed uncertainty and disruption as constants of human life. AI may feel unprecedented, but humans have always had to adapt to major change. Agency matters more than ever You cannot control technological change, but you can control how you respond to it and how you choose to use it. Ryan's website: ryanholiday.net Daily Stoic: dailystoic.com/podcast/ 00:00 Intro: You Can’t Outsource Wisdom 00:29 Meet Ryan Holiday 02:03 The Dream Was To Work Less 03:07 Who Actually Gets The Time? 06:32 Leadership, Culture, And Family First 08:38 How Will You Measure Your Life? 10:11 The Stoic View Of Change 14:44 AI Hallucinations And Shameless Confidence 17:21 You Cannot Outsource Wisdom 19:08 Cognitive Offloading Vs Real Understanding 20:22 Ego, Flattery, And AI 22:52 AI As Editor And Thought Partner 24:59 Mediocre Vs Exceptional Work 31:15 Why Bullshit Detection Matters 38:06 Stoicism, Agency, And Adapting To Change 43:31 The Debrief 📜 Read the transcript for this episode: you-cant-outsource-wisdom-bestselling-author-ryan-holiday-on-what-the-stoics-have-to-say-about-ai/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

April 29, 20261 hr 6 min

Proof of Craft: What It Takes to Stand Out When Everything Looks Good - with Laura Jones, CMO of Instacart

Laura Jones explains that generative AI is raising the bar for creativity. When everyone can produce “pretty good” content, the real challenge is creating something that actually stands out. The risk is not poor output, but settling too quickly for what already works. She argues that as products become more similar, brand becomes a signal of trust. Not in a visual sense, but in the experience behind it. At Instacart, that shows up in details like how a banana is selected. With over a billion bananas delivered and millions of orders including notes on ripeness, customers are expressing very specific preferences. That behavior led to both new product features and the creative idea behind their Super Bowl campaign. The conversation also explores how teams should work with AI. While it can automate repetitive tasks and speed up iteration, it can also create a tendency to agree with what’s generated, especially when working alone. Laura emphasizes that the best ideas still come from people challenging each other, building on different perspectives, and pushing beyond the first acceptable answer. Key takeaways: Mediocre is easier than ever, which raises the bar for originality When AI gets everyone to “pretty good,” the work that stands out has to go further. The bar is not lower. It is higher. Brand becomes trust when products converge As functionality becomes easier to replicate, the question becomes who you trust to get it right. Brand is the answer to that. Only do what only you can do Use AI to take on repetitive work, then spend your time on judgment, insight, and decisions that require a human point of view. Need-finding still requires real people Synthetic research can help, but it cannot replace observing real behavior. The banana insight came from what customers actually did. Human plus bot plus human Working only with AI makes it easy to agree and move on. The best ideas come from people challenging each other, with AI in the middle, not as the whole process. Instacart: instacart.com Super Bowl ad: Super Bowl (Instacart ad) Laura LinkedIn: linkedin/laurajones ro's post: ro.co/perspectives/super-bowl-economics 00:00 Intro: Originality vs AI Complacency 00:27 Meet Laura Jones 01:23 Brand as trust when products converge 03:50 Personalization and reducing mental load 06:24 What still matters in marketing 10:33 Why need-finding cannot be shortcut 14:09 Using AI without losing judgment 16:33 New channels and where customers actually are 21:35 Why “dopey ideas” matter 25:42 Human plus bot plus human 28:44 Inside the Super Bowl ad 31:47 From banana insight to product 34:49 Taking creative risks at scale 37:34 Fear, pressure, and team chemistry 46:24 AI and faster prototyping 53:26 The debrief 📜 Read the transcript for this episode proof-of-craft-what-it-takes-to-stand-out-when-everything-looks-good-with-laura-jones-cmo-of-instacart/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.

April 15, 20261 hr 3 min

Nobody Is Getting New Manager Training for Their AI Team - with Dan Klein, UC Berkeley

Dan Klein, professor at UC Berkeley and CTO at Scaled Cognition, explains that AI systems generate answers based on patterns in language rather than verified knowledge. This makes them highly capable across many tasks, but also means they can produce confident answers even when they are not fully accurate. He introduces the “jagged frontier,” where AI performs very well in some areas and less reliably in others. Because responses are fluent and convincing, it is often hard to see where those limits are, which makes it important to stay engaged when using these systems. The conversation also explores hallucinations as a natural part of generative systems. In some cases, this is what makes them valuable, especially for creative or open-ended tasks, while in other cases reliability becomes more important. Finally, Dan highlights that working effectively with AI is a skill. As more people start using these systems in their daily work, knowing how to guide them, evaluate outputs, and apply them in the right contexts becomes increasingly important. He also shares how his team at Scaled Cognition is tackling this challenge by building AI systems with fundamentally different architectures, focused on determinism and reliability — aiming to ensure systems follow rules, reflect underlying data accurately, and behave predictably in high-stakes, policy-driven use cases. Key Takeaways: AI is designed to sound right, not to know it’s right Models generate fluent answers without knowing whether they are correct, which means users need to actively evaluate outputs You have to learn where AI works and where it doesn’t Capabilities are uneven, and understanding those limits is key to using AI effectively Working with AI shifts your role from creator to editor Instead of starting from scratch, you are reviewing, refining, and validating what the model produces Most people are using AI without knowing how to manage it Skills like delegation, verification, and judgment are becoming essential, but are not widely taught Dan's LinkedIn: linkedin/dan-klein/ Scaled Cognition Website: scaledcognition.com Scaled Cognition LinkedIn: linkedin/company/scaledcognition/ Scaled Cognition X: x.com/ScaledCognition 00:00 Intro: Fluency vs Truth 00:34 Meet Dan Klein 02:53 Why Fluency Misleads 05:11 How LLMs Guess 07:30 What Is Hallucination 08:54 Deception and Alignment 11:22 Why Agents Break 12:48 Chaining and Determinism 16:01 When Hallucination Helps 22:33 Beyond Scale for Reliability 30:40 Synthetic Data Training 31:10 Enterprise Agent Use Cases 33:44 Healthcare Risks 39:13 Enterprise Literacy Gap 41:27 Delegation and AI Management 54:37 The Debrief 📜 Read the transcript for this episode: nobody-is-getting-new-manager-training-for-their-ai-team-with-dan-klein-uc-berkeley/transcript For more prompts, tips, and AI tools. Check out our website: https://www.beyondtheprompt.ai/ or follow Jeremy or Henrik on Linkedin: Henrik: https://www.linkedin.com/in/werdelin Jeremy: https://www.linkedin.com/in/jeremyutley Show edited by Emma Cecilie Jensen.