The Dr. Data Show with Eric Siegel and Luba Gloukhova

In this episode, Eric Siegel narrates his article:

Jev Fever: Predictive AI Is Superhot Again

A new breed of models has rejuvenated predictive AI. But successful deployment demands a new business-value playbook.

... and then Eric summarizes his follow-on article:

Jev AI Will Make You Money – But Exactly How Much?

AI users are pouncing on the new predictive capabilities of Jev and its competitors. But to make money with Jev, you must also "valuate" it for your project.

Listen in full to the new preface to Eric Siegel's The AI Playbook:

Predictive AI Thrives, Despite Generative AI Stealing the Spotlight

This is the new preface to the paperback edition of The AI Playbook: Mastering the Rare Art of Machine Learning Deployment.

Special offer: Pre-order the paperback of The AI Playbook now and receive free, immediate access to the audiobook.

This was also published as an adapted article in:

CDO Magazine: GenAI May Own the Hype, but Predictive AI Also Delivers Value

The Machine Learning Times: GenAI Owns the Hype, but Predictive AI Is Thriving

What'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

In 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

In 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?