Podcast
The Dr. Data Show with Eric Siegel and Luba Gloukhova
Hosted by Eric Siegel · EN
Eric Siegel and Luba Gloukhova cover why machine learning is the most important, most potent, and most misunderstood technology. And did we mention most important?
Yup, it’s the most important – yet most new ML projects fail to deliver value. This podcast will help you:
- Make sure machine learning is effective and valuable
- Catch common machine learning oversights
- Understand ethical pitfalls – concretely
- Sniff out all the ”artificial intelligence” malarky
This podcast is for both data scientists and business leaders of all kinds – such as executives, directors, line of business managers, and consultants – who are involved in or affected by the deployment of machine learning.
To get machine learning to work, both the tech and business sides must make an effort to reach across wide chasm.
About the host:
Eric Siegel, Ph.D., is a leading consultant and former Columbia University professor who helps companies deploy machine learning. He is the founder of the long-running Machine Learning Week conference series, the instructor of the acclaimed online course “Machine Learning Leadership and Practice – End-to-End Mastery,” executive editor of The Machine Learning Times, and a frequent keynote speaker. He wrote the bestselling ”Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die,” which has been used in courses at hundreds of universities, as well as ”The AI Playbook: Mastering the Rare Art of Machine Learning Deployment.” Eric’s interdisciplinary work bridges the stubborn technology/business gap. At Columbia, he won the Distinguished Faculty award when teaching the graduate *computer science* courses in ML and AI. Later, he served as a *business school* professor at UVA Darden. Eric has appeared on numerous media channels, including Bloomberg, National Geographic, and NPR, and has published in Newsweek, HBR, SciAm blog, WaPo, WSJ, and more.
https://www.machinelearningweek.com
http://www.bizML.com
http://www.machinelearning.courses
http://www.thepredictionbook.com
11episodes
Episodes
Newest firstAll episodes
The Doomer's Error: Why AGI Is An Incoherent Concept
Mar 2500:48:29Tap to summarizeWhat's the strongest anti-AGI case, the argument that reveals the fallacies underlying the belief that AGI is a viable goal – as well as the AI doomerism that believing AGI will soon arrive often spawns? Princeton professor Arvind Narayanan recently made a statement that we feel deserves amplification: For real-world problems, machines face some of the same key fundamental limits and challenges that humans face. Listen to Luba and Eric unpack, explore, and expound. #noAGI
Transcribe →Predictive AI vs. GenAI: A Crucial, Unavoidable Comparison
Mar 1700:48:17Tap to summarizeIn this episode we cover: - Why predictive AI and generative AI are destined to remain inherently distinct - Why comparing them is unavoidable, even though they solve different problems - How they compare - How companies should balance investments between the two
Transcribe →Pushing the ultimate limits: helping genAI realize its promise of autonomy
Mar 1100:53:56Tap to summarizeIn this episode, we talk about real, truly deployed LLM-based systems that push the limits of autonomy. How can we "tame" LLMs to create feasible, practical solutions that are viable for deployment? What are their ultimate limitations?
Transcribe →Superhuman Adaptable Intelligence: LeCun's New Buzzword Challenges AGI
Mar 500:55:51Tap to summarizeIn this episode, Luba Gloukhova and Eric Siegel unpack the new paper, "AI Must Embrace Specialization via Superhuman Adaptable Intelligence," by Yann LeCun and others. The paper endeavors to "address what’s wrong with our conception of AGI, and why, even in its most coherent formulation, it is a flawed concept to describe the future of AI." That aligns so well with our episode just two days ago that one of the paper's authors, Philippe Wyder, tweeted us about the paper, bringing it to our attention! The paper presents the new term "Superhuman Adaptable Intelligence," which is defined as "intelligence that can learn to exceed humans at anything important that we can do, and that can fill in the skill gaps where humans are incapable." Listen to our break-down and take, and access the full paper here: https://arxiv.org/abs/2602.23643
Transcribe →The Whole Problem with AGI and Its Ridiculous Definitions
Mar 300:40:11Tap to summarizeIn this very special episode, the first with a co-host (Luba Gloukhova), Dr. Data and Miss Information explore why people are messing with the definition of artificial general intelligence, the problem with the concept, how it feeds AI hype, and how we can feasibly realize a good portion of genAI's overzealous promise of autonomy.
Transcribe →Predictive AI Thrives, Despite GenAI Stealing The Spotlight
Feb 1100:10:21Tap to summarizeIn this episode, listen to a narration of Eric Siegel's article in Forbes: Predictive AI Thrives, Despite GenAI Stealing The Spotlight GenAI and predictive AI battle for resources, but even as the overwhelming attention focuses on genAI, enterprises are still adopting predictive AI just as much. Access the original article here: https://www.forbes.com/sites/ericsiegel/2026/02/11/predictive-ai-thrives-despite-genai-stealing-the-spotlight/ You can access an overview of HYBRID AI 2026 and a description of each enterprise presentation here: https://machinelearningweek.com/
Transcribe →Hybrid AI: Industry Event Signals Emerging Hot Trend (article)
Feb 900:11:04Tap to summarizeIn this episode, listen to a narration of Eric Siegel's article in Forbes: Hybrid AI: Industry Event Signals Emerging Hot Trend AI is not yet the success that it should be, so two dozen enterprises will disclose their move toward a crucial new paradigm – hybrid AI – at a 2026 conference. Access the original article here: https://www.forbes.com/sites/ericsiegel/2026/02/09/hybrid-ai-industry-event-signals-emerging-hot-trend/
Transcribe →A Gooder AI case study: profiting on machine learning
Sep 300:33:55Tap to summarizeThe biggest hurdle for data science teams isn't building the model; it's proving its dollar value. This presentation shows how a dental group could translate a no-show prediction model into a clear business case worth $$$ It's about shifting the conversation from abstract metrics to tangible ROI. Henry Castellanos is a data scientist extraordinaire. He goes beyond establishing a strong technical performance for his ML models to also maximizing their business value. Let this sink in: Most data scientists don't do that – most ML projects don't plan and sell predictive AI deployment according to the the explicit business value. Interestingly, Henry points out that using Gooder AI (www.gooder.ai) to do this even bucks up his own confidence in his models and their business value. Listen to Henry's presentation to see exactly how to bridge the gap from ML to real-world value. To view this presentation as a video, go to: https://youtu.be/BT-GnnuN3jA
Transcribe →Predictive AI Usually Fails Because It’s Not Usually Valuated (article)
May 5, 202500:03:46Tap to summarizeIn this episode, listen to a narration of Eric Siegel's article in Forbes: Predictive AI Usually Fails Because It’s Not Usually Valuated Most predictive AI deployments are scrubbed. Why? They didn't forecast the potential value in business terms like profit or savings. Access the original article here: https://www.forbes.com/sites/ericsiegel/2024/11/18/predictive-ai-usually-fails-because-its-not-usually-valuated/
Transcribe →Predictive AI Only Works If Stakeholders Tune This Dial (article)
Apr 14, 202500:04:26Tap to summarizeIn this episode, listen to a narration of Eric Siegel's article in Forbes: Predictive AI Only Works If Stakeholders Tune This Dial Machine learning models can drive business operations to great benefit. But, to get there, stakeholders must determine how model probabilities trigger actions. Access the original article here: https://www.forbes.com/sites/ericsiegel/2024/11/25/predictive-ai-only-works-if-stakeholders-tune-this-dial/
Transcribe →