Daniel Guetta on the Guts of AI, Agentic AI & Why LLMs Hallucinate | The Real Eisman Playbook Ep 46
TL;DR
Columbia Business School Professor Daniel Guetta explains that Large Language Models are fundamentally probabilistic autocomplete engines trained on internet data, converting words to numerical embeddings to predict sequences—making hallucinations an inherent feature rather than a malfunction when handling novel information.
🧠 AI Fundamentals: Two Distinct Paradigms 2 insights
Predictive AI dominates current value
Traditional machine learning (predictive AI) has generated business value since the 1980s through numerical pattern recognition for credit scoring, fraud detection, and pricing models like Zillow's Zestimate, which tweaks parameters against historical sales data.
Generative AI unlocks unstructured data
Unlike predictive AI which requires Excel-style numerical inputs, GenAI and Large Language Models process text and images through deep neural networks, overcoming the limitation of earlier systems that couldn't multiply text or analyze visual content.
⚙️ The Mechanics of LLMs 3 insights
Sophisticated autocomplete engines
LLMs generate responses by predicting one word at a time based on probability, sequentially building sentences by determining what word should follow the previous sequence, requiring the model to reprocess the entire conversation history for each new token.
Embeddings convert language to mathematics
Words are transformed into numerical vectors (embeddings) with thousands of dimensions based on co-occurrence patterns in training data, causing conceptually similar words like 'king' and 'queen' to cluster together while unrelated terms like 'Pringle' and 'existentialism' remain distant.
Internet-scale training creates the model
Models learn these embedding values through machine learning algorithms that ingest vast portions of internet text, automatically adjusting numerical weights until the statistical patterns match the training corpus without human-defined rules.
⚠️ Why Hallucinations Are Inevitable 2 insights
Pattern mimicry versus factual retrieval
LLMs hallucinate because they statistically predict plausible-sounding word sequences rather than retrieving verified facts, meaning they should be expected to generate incorrect information when the statistical patterns in their training data don't reflect reality.
The novelty problem
These models fail catastrophically with breaking news or novel events—such as when ChatGPT incorrectly stated Maduro was still in Venezuela immediately after he was removed—because they rely entirely on historical patterns absent from their training data.
Bottom Line
Treat LLMs as probabilistic pattern-matching autocomplete tools rather than knowledge databases, implementing verification workflows for factual accuracy while recognizing that their probabilistic nature makes hallucinations unavoidable, particularly with real-time or novel information.
More from Steve Eisman
View all
The Only Health Insurance Stock You Can Own with Michael Ha | The Real Eisman Playbook Episode 51
Baird analyst Michael Ha reveals how United Healthcare's Optum Health division faces structural collapse due to new Medicare risk adjustment rules exposing years of aggressive patient coding, while delusional investors cling to the hope that traditional insurance repricing can offset permanent revenue impairment.
Mark Cuban Shares His Plan For Fixing Drug Prices in America | The Real Eisman Playbook Ep 50
Mark Cuban details how his company Cost Plus Drugs undercuts the pharmaceutical industry by selling medications at cost plus 15%, while exposing how Pharmacy Benefit Managers (PBMs) force the sickest Americans to pay inflated prices that subsidize employer rebates and premiums.
The Iran War Crisis: Ex-CIA Analyst Helima Croft on Oil Shock & Energy Risk | The Weekly Wrap
Former CIA analyst Helima Croft warns that the Iran war has triggered a genuine energy crisis with the Strait of Hormuz effectively closed, 6.7 million barrels of daily production shut in due to full storage tanks, and depleted Western strategic reserves insufficient to cover more than a few weeks of disruption.
Private Credit Cracks, The AI "Boogeyman" & Why Crypto is for Boomers | The Real Eisman Playbook
Steve Eisman, Glenn Shore, and Dwight Collins dissect the recent panic in private credit markets, focusing on Blue Owl's failed BDC restructuring and the structural relationship between direct lending and private equity, while warning that alternative asset managers have become uninvestable due to complexity and headline-driven retail outflow risks.