New Lex Fridman Insight: Michael I. Jordan: Machine Learning, Recommender Systems, and Future of AI
Sent June 11, 2026
Key Insights
- Michael I. Jordan argues AI is still a proto-field, akin to early chemical engineering, not yet achieving true intelligence.
- Jordan critiques the term 'AI' as misleading, advocating for 'machine learning' to better reflect the field's current capabilities.
- Recommender systems, though not magical, have become a billion-dollar industry crucial for consumer markets.
- Jordan emphasizes decision-making over prediction in AI, challenging the notion that AI's primary value lies in predictive accuracy.
- Jordan highlights the intelligence of markets as a distinct form of intelligence, separate from human cognition.
How the conversation moved
The host begins by framing the conversation around the evolution of AI, asking Michael I. Jordan to compare its current state to other engineering fields. Jordan sets the stage by describing AI as a proto-field, akin to the early days of chemical and electrical engineering, emphasizing that true intelligence, as envisioned by early AI pioneers, remains an aspiration rather than a reality. He argues that the current focus on AI is more about data processing and machine learning than genuine intelligence, suggesting that the field is still in its infancy and has a long way to go before it can achieve the ambitious goals set by its founders.
Jordan's main argument centers on the misleading nature of the term 'AI', which he believes has led to inflated expectations and misconceptions about the field's capabilities. He provides concrete evidence by highlighting the economic impact of machine learning applications, such as recommender systems, which have become a billion-dollar industry. These systems, he argues, demonstrate the practical applications of machine learning, but they do not reflect the kind of intelligence that the term 'AI' implies. Jordan suggests that a more accurate terminology would help align public and academic perceptions with the field's actual progress and potential.
The host challenges Jordan's framing by suggesting that AI's current capabilities might still represent a form of intelligence, albeit different from human cognition. Jordan pushes back, asserting that the term 'AI' is a misnomer that has led to unrealistic promises, particularly in areas like brain-computer interfaces and understanding human cognition. He argues that the field should focus more on decision-making processes rather than prediction, as decision-making is where AI can have the most significant real-world impact. This pushback highlights a tension between the aspirational goals of AI and its current practical applications.
In the resolution, Jordan reiterates his belief that the field's focus should shift towards understanding and improving decision-making processes, which he sees as more achievable and impactful than striving for perfect prediction. The conversation pivots to discuss the intelligence of markets, which Jordan describes as a distinct form of intelligence separate from human cognition. He concludes by emphasizing the importance of developing new terminologies and frameworks that accurately reflect the field's capabilities and potential, suggesting that this shift could lead to more meaningful advancements in AI and related disciplines.
Surprising moments
In-depth
AI as a Proto-Field
- Jordan likens AI to early chemical engineering, suggesting it's still in its infancy.
- He argues current AI lacks true intelligence, focusing more on data processing.
Critique of 'AI' Terminology
- Jordan argues 'AI' is a misnomer, advocating for 'machine learning' as a more accurate term.
- He believes the term 'AI' sets unrealistic expectations.
Economic Impact of Recommender Systems
- Recommender systems are a billion-dollar industry, crucial for modern consumer markets.
- Effective recommender systems significantly impact sales and consumer engagement.
Decision-Making in AI
- Jordan emphasizes decision-making over prediction in AI applications.
- He argues that focusing on decision-making could lead to more practical AI solutions.
Market Intelligence
- Jordan describes markets as intelligent systems distinct from human intelligence.
- He suggests studying market intelligence could provide new economic insights.
Still open
- Jordan questions whether current AI can ever achieve the level of intelligence originally envisioned by its pioneers.
- The host asks how the field of AI might evolve if it shifts focus from prediction to decision-making.