Artificial Intelligence Snake Oil: What AI Can and Can't Do - Featuring Professor Arvind Narayanan | Episode #0010
Start Small, Think Big - A podcast and newsletter that guides you on your AI and Automation journey.Show Notes
Follow the SSTB Newsletter.
Summary In this episode of the Start Small Think Big podcast, the hosts discuss the complexities and nuances of artificial intelligence, particularly focusing on the concept of 'AI snake oil'—the misleading promises made by some AI solutions. They explore current events in the tech industry, the ethical implications of AI in hiring practices, and the importance of human oversight in AI applications. The conversation also touches on the cultural shift regarding AI, including its recognition by influential figures like the Pope. The hosts emphasize the need for discernment in AI deployment, the human element in AI training, and the importance of asking the right questions when considering AI solutions.
Keywords AI, artificial intelligence, ethics, hiring, employment, technology, generative AI, human oversight, predictive analytics, business decisions
Takeaways AI is a nuanced topic that requires careful consideration. Current events in tech highlight the rapid changes in the industry. Ethical implications of AI in hiring practices are significant. Human oversight is crucial in AI applications to avoid errors. Generative AI can be a powerful tool but comes with risks. Cultural shifts regarding AI are being recognized at high levels. Understanding AI's diverse applications is essential for businesses. The importance of discernment in AI deployment cannot be overstated. Asking the right questions is key to effective AI implementation. Failure can lead to valuable learning experiences in AI development.
Titles Unpacking AI Snake Oil: What You Need to Know The Ethics of AI: Navigating Hiring Practices
Sound Bites "AI is such a nuanced topic." "Trust but verify." "Don't be afraid to fail."
Chapters 00:00 Introduction to AI Snake Oil 01:26 The Impact of AI on Employment 04:32 Misrepresentation of AI Capabilities 08:06 The Importance of Human Oversight 11:42 Navigating Job Applications in the AI Era 15:13 Cultural Shifts in AI Acceptance 18:58 The Role of AI in Professional Development 22:29 The Pope and AI: A Unique Perspective 25:28 AI's Societal Impact and Recognition 27:24 Understanding AI Snake Oil 28:14 Diverse Nature of AI 30:12 Navigating AI Implementation 33:17 The Role of LLMs in Business 36:57 Predictive AI in High-Stakes Areas 40:30 Generative AI: Promise and Perils 47:45 Human Element in AI Training 52:45 The Importance of Discernment in AI Deployment 53:29 Internal Communication and Stakeholder Engagement 55:28 Assessing AI Effectiveness and Application 59:20 Steering AI Towards Positive Outcomes 01:03:27 Caution Against Extreme Opinions in AI 01:06:14 Identifying Hype and Misleading Claims 01:11:15 The Reality of Predictive Analytics and Data Capture 01:16:33 Final Thoughts on Responsible AI Implementation
https://linktr.ee/digitalmeld We build tools that save businesses time and money by helping them unlock their greatest potential and achieve growth through data, automation, and AI-powered solutions.
More from Start Small, Think Big - A podcast and newsletter that guides you on your AI and Automation journey.
AI can help you finish the work. It cannot take responsibility for whether the work is right. Doug Gabbard, Principal Cloud Solution Architect at Microsoft and the creator behind Practicing Photography, joins Brad Groux and Robert Groux to talk about using AI where it actually helps: learning unfamiliar creative tools, finding buried knowledge, improving scripts, and turning experience into better requirements. Doug shares how he uses screenshots to get help in Photoshop and Lightroom, builds agents around departmental SharePoint knowledge, and brings engineering discipline to his photography workflow. We also get into the limits: confident wrong answers, missing security requirements, the future of entry-level learning, and why people skills still matter. The question to take back to your next project: What did I miss? Takeaways Start with a real task you understand, not a tool you want an excuse to use. Give AI the relevant context: source material, desired outcome, constraints, and what failure looks like. A script that runs is not automatically secure, scalable, or ready for other people to use. Ask for criticism and missing requirements, then check the response against evidence. Use AI to repurpose your own ideas without handing over your point of view. Keep learning the fundamentals—and learn to explain them to people. Resources discussed Microsoft Adobe Photoshop Adobe Lightroom ChatGPT Gemini Claude Microsoft Copilot SharePoint Backblaze Azure Blob Storage Synology n8n Supabase OpenClaw PowerShell Microsoft Sentinel Microsoft Agent Builder quick start (PDF) Follow Doug Practicing Photography YouTube LinkedIn Follow Start Small, Think Big Find the podcast, newsletter, and our other channels: https://linktr.ee/digitalmeld More episodes and articles: https://digitalmeld.io/podcast/ Recorded February 19, 2026. Product availability, pricing, and policies discussed reflect the conversation at that time. Doug shares his own experience; this is not an official Microsoft product announcement. Recorded and edited using Riverside — https://go.sstb.ai/riverside (affiliate link). Affiliate disclosure: Some links above are affiliate links. Digital Meld may earn a commission if you purchase through them, at no additional cost to you.
Follow the Start Small, Think Big newsletter: https://www.linkedin.com/newsletters/7305986878653022210/
Follow the SSTB Newsletter. Special thanks to Allen Martinez , and check out BXAIOS.
Follow the SSTB Newsletter. Special thanks to Matt Dorman, and check out NDEVR.
Follow the SSTB Newsletter. Special thanks to Randy Gage.