Skip to content
Lex Fridman PodcastLex Fridman Podcast

Michael I. Jordan: Machine Learning, Recommender Systems, and Future of AI | Lex Fridman Podcast #74

Michael I Jordan is a professor at Berkeley, and one of the most influential people in the history of machine learning, statistics, and artificial intelligence. He has been cited over 170,000 times and has mentored many of the world-class researchers defining the field of AI today, including Andrew Ng, Zoubin Ghahramani, Ben Taskar, and Yoshua Bengio. This episode is presented by Cash App. Download it & use code "LexPodcast": Cash App (App Store): https://apple.co/2sPrUHe Cash App (Google Play): https://bit.ly/2MlvP5w PODCAST INFO: Podcast website: https://lexfridman.com/podcast Apple Podcasts: https://apple.co/2lwqZIr Spotify: https://spoti.fi/2nEwCF8 RSS: https://lexfridman.com/feed/podcast/ Full episodes playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4 Clips playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41 EPISODE LINKS: (Blog post) Artificial Intelligence -- The Revolution Hasn’t Happened Yet: https://hdsr.mitpress.mit.edu/pub/wot7mkc1 OUTLINE: 0:00 - Introduction 3:02 - How far are we in development of AI? 8:25 - Neuralink and brain-computer interfaces 14:49 - The term "artificial intelligence" 19:00 - Does science progress by ideas or personalities? 19:55 - Disagreement with Yann LeCun 23:53 - Recommender systems and distributed decision-making at scale 43:34 - Facebook, privacy, and trust 1:01:11 - Are human beings fundamentally good? 1:02:32 - Can a human life and society be modeled as an optimization problem? 1:04:27 - Is the world deterministic? 1:04:59 - Role of optimization in multi-agent systems 1:09:52 - Optimization of neural networks 1:16:08 - Beautiful idea in optimization: Nesterov acceleration 1:19:02 - What is statistics? 1:29:21 - What is intelligence? 1:37:01 - Advice for students 1:39:57 - Which language is more beautiful: English or French? CONNECT: - Subscribe to this YouTube channel - Twitter: https://twitter.com/lexfridman - LinkedIn: https://www.linkedin.com/in/lexfridman - Facebook: https://www.facebook.com/LexFridmanPage - Instagram: https://www.instagram.com/lexfridman - Medium: https://medium.com/@lexfridman - Support on Patreon: https://www.patreon.com/lexfridman

Lex FridmanhostMichael I. Jordanguest
Feb 24, 20201h 45mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Michael I. Jordan Redefines AI: From Hype to Human-Centric Engineering

  1. Michael I. Jordan argues that what is called “AI” today is not artificial intelligence in the McCarthy sense, but the early stages of a new engineering discipline built on statistics, computation, and economics. He stresses how little we understand about human brains and real intelligence, criticizing hype around brain-inspired AI and short-term claims about human-level language understanding. Much of the conversation centers on decision-making at scale, markets, and recommender systems, contrasting prediction from data with the harder problem of building consequential, human-aligned systems. Jordan also critiques advertising-based business models, calls for producer–consumer markets that genuinely create value, and frames the future of AI as “intelligent infrastructure” that augments rather than replaces humans.

IDEAS WORTH REMEMBERING

5 ideas

Reframe AI as a new engineering discipline, not imminent human-level intelligence.

Jordan likens today’s AI to early chemical or electrical engineering: we’re building large-scale systems from statistical and computational ideas, but we are nowhere near understanding, let alone reproducing, human intelligence. Treating this as engineering clarifies goals and reduces misleading hype.

Prioritize decision-making and markets over pure prediction from data.

He argues the field is over-focused on pattern recognition and prediction (e.g., deep learning demos) and under-focused on decision-making where risk, feedback, externalities, and incentives matter—exactly where real-world consequences for humans arise.

Build real producer–consumer markets that create livelihoods, not just clicks.

Using music as an example, Jordan proposes platforms that transparently connect creators and listeners, enabling thousands of mid-level careers via data dashboards and fair transactions, instead of keeping most economic value with labels or streaming intermediaries.

Advertising-centric monetization structurally distorts platforms and fosters “fake news.”

Optimizing for click-through and ad revenue incentivizes engagement hacks and sensational content; Jordan believes platforms should gradually reduce low-quality ads and replace them with transaction-based revenues where users willingly pay for real value.

Respect human agency and context in recommender systems and privacy.

He sees current recommender systems as overreliant on passive behavioral traces and opaque profiling; instead, systems should be transparent, give users control over when and how they’re guided, and support discovery without creeping surveillance.

WORDS WORTH SAVING

5 quotes

I think what’s happening right now is not AI. That was an intellectual aspiration. What we have is the emergence of a new engineering discipline based on statistics and computation.

Michael I. Jordan

We have no clue how the brain does computation. We’re like the Greeks speculating about going to the moon.

Michael I. Jordan

Prediction plus decision-making is everything, but both of them are equally important. The field has emphasized prediction at the expense of decision-making, where human lives are at stake.

Michael I. Jordan

Advertising has completely taken over the business model. Click‑through rate is the core problem. You’ve got to remove that if you want to fix fake news.

Michael I. Jordan

An engineering discipline can be what we want it to be. In the current era we have a real opportunity to conceive of something historically new, a human‑centric engineering discipline.

Michael I. Jordan (quoted by Lex Fridman from Jordan’s essay)

Distinction between classical AI aspirations and modern machine learning/engineeringOur limited scientific understanding of the brain and human intelligencePrediction vs. decision-making under uncertainty in real-world systemsMarkets, platforms, and recommender systems as forms of large-scale intelligenceFailures and risks of advertising-driven business models (Google, Facebook, etc.)Privacy, control, and human-centric design in data-driven systemsFoundations in optimization, statistics, Bayesian vs frequentist thinking, and game theory

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome