A Beginner's Guide to AI

Humayun Sheikh on the Agentic Web, Trust, and the Agentic EconomyHumayun Sheikh joins Dietmar Fischer to explain what happens when AI stops recommending and starts doing. We explore the Agentic Web, a new layer where personal AI agents and verified brand agents collaborate to complete tasks like booking travel, coordinating meetings, and shopping with trust built in.You will learn what makes a real AI agent, why autonomy matters, and how multi-agent systems unlock an agentic economy. We also tackle the marketer’s question: what happens to SEO when the buyer becomes an assistant agent choosing on your behalf? Humayun breaks down how identity, verification, and trusted lists can reduce scams and make agentic commerce safe and usable.📧💌📧Tune in to get my thoughts and all episodes, don't forget to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠subscribe to our Newsletter⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠: ⁠⁠⁠⁠beginnersguide.nl⁠⁠⁠⁠📧💌📧About Dietmar FischerDietmar is a podcaster and AI marketer from Berlin. If you want to know how to get your AI or your digital marketing going, just contact him at argoberlin.comChapters00:00 Welcome and Humayun’s journey from gaming to DeepMind03:01 What is an AI agent: autonomy and decision-making08:20 The Agentic Web: discoverability, connectivity, trust and commerce rails23:47 Personal agents in practice: preferences, handles and onboarding in minutes29:53 Verified brand agents and trust: domains, identity and safe agentic buying48:12 Risks, AGI fears, corporations vs countries and what comes nextQuotes from the Episode“There has to be a hint of autonomy within an agent.”“We have provided the rails of discoverability, connectivity, communication, trust. And commerce.”“Your aggregator is your own agent. It holds your preferences. It doesn’t pass it to anybody.”“Anybody who has a website should have an agent, or will have an agent.”“I was the first investor in DeepMind.”“We will not have countries, we will have corporations.”Where to find Humayun SheikhFetch.ai - your personal AIASI1.ai - the LLMFollow Humayun on LinkedIn!Music credit: "Modern Situations" by Unicorn Heads Hosted on Acast. See acast.com/privacy for more information.

The Rising Cost of Intelligence: What Expensive AI Means for the WorldArtificial intelligence is reshaping how we work, learn, and create. But as frontier AI models become more capable, their costs are rising faster than ever. This episode of A Beginner’s Guide to AI dives into the global AI divide, exploring how price, compute, infrastructure, and access are quietly determining who benefits from AI and who risks falling behind.Listeners will discover why advanced AI models cost so much to train and run, how high prices can concentrate innovation in wealthy institutions, and why access to strong models is becoming a new form of economic and educational inequality. Through vivid examples and clear explanations, Professor Gephardt guides listeners through the real-world consequences of expensive AI and what can still be done to ensure a more inclusive future.📧💌📧Tune in to get my thoughts and all episodes, don't forget to subscribe to our Newsletter: beginnersguide.nl📧💌📧About Dietmar Fischer Dietmar is a podcaster and AI marketer from Berlin. If you want to know how to get your AI or your digital marketing going, just contact him at argoberlin.comQuotes from the Episode:“When intelligence becomes expensive, opportunity becomes exclusive.”“A great model is useless if only a handful of people can afford to use it.”“If AI becomes a privilege, innovation shrinks to the size of the elite who control it.”Chapters00:00 The Hidden Price of Intelligence04:12 Why Cutting-Edge AI Is So Expensive12:47 How AI Costs Create a Global Divide21:30 Real-World Case Studies on AI Access32:18 Practical Ways to Narrow the AI Gap39:42 Final Thoughts and Key LessonsMusic credit: "Modern Situations" by Unicorn Heads 🎧✨ Hosted on Acast. See acast.com/privacy for more information.

Context rot is one of the most underestimated risks in artificial intelligence today. In this episode of A Beginner’s Guide to AI, we explore how AI systems trained on static data slowly drift away from reality while continuing to sound confident, helpful, and persuasive.You’ll learn why large language models struggle with time, why feeding more information into AI can backfire, and how outdated knowledge quietly sabotages decisions in marketing and business. This episode explains the difference between timeless principles and perishable insights, and why trusting AI without checking freshness can cost credibility and money.Key topics include context rot in AI, outdated training data, long context window limitations, AI decision-making risks, and practical strategies like retrieval-augmented generation and smarter context engineering.📧💌📧Tune in to get my thoughts and all episodes, don't forget to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠subscribe to our Newsletter⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠: beginnersguide.nl📧💌📧About Dietmar Fischer:Dietmar is a podcaster and AI marketer from Berlin. If you want to know how to get your AI or your digital marketing going, just contact him at argoberlin.comQuotes from the Episode“Fluency is not accuracy, even though our brains desperately want it to be.”“More context doesn’t make AI smarter, it often makes it confused.”“AI confidence is cheap. Verification is expensive.”Chapters00:00 Context Rot and the Illusion of Smart AI05:42 Why AI Knowledge Freezes in Time12:18 When More Context Makes AI Worse19:47 Business and Marketing Risks of Context Rot27:05 How to Reduce Context Rot in Practice34:40 What Humans Must Do Better Than AIMusic credit: "Modern Situations" by Unicorn Heads 🎧 Hosted on Acast. See acast.com/privacy for more information.

Machine learning is everywhere, yet rarely understood. In this episode of A Beginner’s Guide to AI, we strip away the hype and explain how machine learning actually works, why it’s so powerful, and where it quietly goes wrong.You’ll learn how machines are trained on data rather than rules, why predictions are not understanding, and how real-world systems can produce unfair outcomes even when they look accurate. A real healthcare case shows how a cost-based algorithm systematically underestimated medical need, revealing the hidden dangers of proxy metrics.This episode covers machine learning basics, ethical AI, algorithmic bias, fairness, and transparency in a way that is accessible to beginners and useful for professionals.📧💌📧Tune in to get my thoughts and all episodes, don’t forget to subscribe to our Newsletter: beginnersguide.nl📧💌📧Quotes from the Episode“Machine learning gives you what you measure, not what you value.”“The algorithm didn’t invent bias. It learned it efficiently.”“A perfect prediction of the wrong thing is still failure.”Chapters00:00 Machine Learning Without the Myth04:12 How Machines Learn From Data10:45 Types of Machine Learning18:30 The Cake Example26:05 Healthcare Case Study36:40 Ethics, Bias, and Proxies45:50 Final TakeawaysAbout Dietmar Fischer: Dietmar is a podcaster and AI marketer from Berlin. If you want to know how to get your AI or your digital marketing going, just contact him.Music credit: Modern Situations by Unicorn Heads Hosted on Acast. See acast.com/privacy for more information.

REPOST due to low podcast listener activity - if you listen now, you are the exception 😉Ever wondered how Netflix knows exactly what you'll binge next or how big brands like Delta Air Lines turn multimillion-dollar sponsorships into concrete sales?Welcome back to A Beginner's Guide to AI, where today we're uncovering the fascinating world of AI inference—the secret sauce behind machine-made predictions.--- --- ---A word from our Sponsor:Sensay creates AI-powered digital replicas to preserve and share individual and organizational knowledge, turning it into scalable, sustainable, and autonomous wisdom.Visit Sensay at ⁠⁠⁠⁠⁠⁠⁠Sensay.io⁠⁠⁠⁠⁠⁠⁠And listen to Dan, Sensay's CEO and founder, ⁠⁠⁠⁠⁠⁠⁠in this episode⁠⁠⁠⁠⁠⁠⁠!--- --- ---Professor Gephardt, with his usual charm and wit, breaks down precisely how AI learns from past data to tackle new, unseen scenarios, turning educated guesses into powerful, profitable insights.Expect engaging analogies—from fruit-loving robots to cake-tasting mysteries—and real-life case studies, like Delta’s remarkable $30 million Olympic success story powered by AI. Plus, practical tips on how to spot AI inference in your daily digital life and even how to experiment with your own AI models!Tune in to get my thoughts, and don't forget to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠subscribe to our Newsletter⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠!This podcast was generated with the help of ChatGPT and Mistral. We do fact-check with human eyes, but there still might be hallucinations in the output. And, by the way, it's read by an AI voice.Music credit: "Modern Situations" by Unicorn Heads Hosted on Acast. See acast.com/privacy for more information.