Connected & Protected
24/7 Support
HomeBlogDon’t be fooled—LLMs don’t reason
Artificial intelligence
#Artificial intelligence
#App
#Opinion
#large language models
#AI reasoning capabilities
#AlphaGo
#machine intuition vs reasoning
#AI decision-making
#cognitive processes in AI
#Deep Blue vs AlphaGo
#human-like reasoning in AI
#artificial intelligence insights
#complex decision-making in AI
#Go game AI strategies
#future of AI technology

Don’t be fooled—LLMs don’t reason

On an afternoon in Seoul in March 2016, I watched a program I helped build put a stone on the fifth line of a Go board in what looked like a gift to its human opponent. Move 37 in game two of the five-game match looked so absurd that some commentators thought it was a…

T
Thore Graepel
Content Author
Content Specialist

Thore Graepel is a leading expert in artificial intelligence, specializing in machine learning and its applications in strategic decision-making. With a rich background in cognitive science, he explores the intersection of AI and human reasoning.

Expertise: Artificial intelligence, machine learning, cognitive science, strategic decision-making, AI ethics, human-computer interaction, neural networks
Published: October 6, 2026
8 min read

Quick Answer

Large language models (LLMs) may excel in generating text but lack true reasoning capabilities. This post examines AlphaGo's decision-making process versus LLM operations.

TL;DR

This article discusses the limitations of large language models (LLMs) in reasoning, contrasting them with AlphaGo's complex decision-making during its historic Go match. It highlights the need for AI systems to develop genuine reasoning capabilities for reliable insights.

Key Takeaways

  • 1AlphaGo's move 37 reflects machine reasoning, not mere intuition.
  • 2Current LLMs rely on token prediction, lacking true decision-making processes.
  • 3Future AI advancements require integrating reasoning capabilities similar to human cognition.
Don’t be fooled—LLMs don’t reason

Practical how-to • For business readers

How to Understand AI's Limitations in Reasoning

Small business owners can benefit from recognizing AI's reasoning limits.

  1. Identify AI's reasoning limits

    Recognize that current AI models, like large language models, primarily operate on fast, intuitive responses rather than true reasoning capabilities, as highlighted in the article.

  2. Use AI cautiously

    When using AI tools for decision-making, remember that they may not provide reliable reasoning or insights, especially in critical areas like medical or scientific applications.

  3. Evaluate AI-generated outputs

    If you receive an AI-generated suggestion or conclusion, assess its validity by checking the underlying evidence or reasoning provided, as AI often lacks a clear audit trail.

Guidance prepared on October 6, 2026 • Source published October 2, 2026. Based on the article and its source; check with your IT team before changing work systems.

This guide may be out of date. Check the linked source and current instructions before acting.

On an afternoon in Seoul in March 2016, I watched a program I helped build put a stone on the fifth line of a Go board in what looked like a gift to its human opponent. Move 37 in game two of the five-game match looked so absurd that some commentators thought it was a programming glitch.

It wasn’t. AlphaGo won the game, ultimately triumphing 4-1 over Lee Sedol, one of the greatest professional Go players of all time. “I thought AlphaGo was based on probability calculation and that it was merely a machine,” Lee said afterwards. “But when I saw this move, I changed my mind. Surely, AlphaGo is creative.” 

When Deep Blue defeated then reigning world chess champion Garry Kasparov in 1997, it did so by looking six to eight moves ahead per player and evaluating 200 million chess positions per second, using rules hard-coded by humans. Go is a vastly more complex game. A stone’s worth depends on how distant groups and territory unfold over dozens of moves. Computing even a fraction of the possible outcomes would take a supercomputer billions of years. To win, AlphaGo had to sense who was ahead at a glance and even invent moves no human had thought to play.

That is why many accounts of AlphaGo’s match against Lee portray move 37 as a flash of pure machine intuition. But that is a misunderstanding. It was actually AlphaGo’s powers of reasoning that made this creative choice—and these are powers that today’s AI lacks. If we want future AI systems to produce trustworthy results and really novel insights in fields like science and medicine, we need to equip them with genuine reasoning capabilities of this kind.

AlphaGo is made up of two systems. The first, its policy network, was trained to guess what move a strong human would play. This “intuitive” part regarded move 37 as nothing special—a play that had a roughly one in 10,000 chance of being made by an expert human player. What made AlphaGo choose it was the program’s search machinery, which looked beyond immediate plausibility and weighed the future consequences of proposed moves. It explicitly constructed and searched a game tree with thousands of branches, each representing a different possible future.&nbs

Acronis
Related Product

Acronis Cyber Backup Cloud Advanced - VM (per Device)

by Acronis

Acronis Cyber Backup Cloud Advanced - VM (per Device) - Enterprise-grade backup & recovery solution

Learn More
p;

A well-known theory in the behavioral sciences, popularized by Daniel Kahneman, distinguishes between two modes of human thought: System 1 is fast, gut-level, effortless; system 2, slow, step-by-step, and deliberative. AlphaGo offered a striking machine analogue of that split. Its networks supplied the hunches—this move looks promising, this position looks won—and its search supplied the deliberation, testing those hunches against the moves and countermoves that would follow. As in human cognition, neither half works alone. Intuition alone would never have opted for move 37, and brute-force search would have struggled to sieve through all the many possible moves.

This is strikingly different from the way today’s AI models work. A large language model picks the next token, over and over. That amounts to system 1 in action—fast, associative, and surprisingly good pattern completion across almost every subject people write about.

Not long after ChatGPT debuted, the field realized that language fluency alone falls short of true usefulness. The apparent solution was to make models that deliberate: Instead of answering immediately, they can now generate intermediate steps that decompose a problem, carry forward partial results, and influence subsequent reasoning—a process known as chain of thought.

The gains have proved real, above all in mathematics and coding. But unlike AlphaGo’s search, this does not introduce a genuinely separate reasoning mechanism: The intermediate reasoning is still produced by the same next-token prediction process, iterated for longer before the model commits to an answer.

Three shortcomings prevent what chatbots do from qualifying as reasoning (in a way that a scientist might recognize). First, these models typically maintain no explicit, persistent, and inspectable epistemic state. There is no open ledger that lays out the hypotheses a model is considering, the confidence it has in various explanations, the evidence it’s weighing, and the unresolved questions it’s holding onto—all things that should be systematically revised as new information arrives.

Second, they lack a clean separation between what the system knows and how it manipulates that knowledge. Knowledge and reasoning are inextricably interwoven in the weights of the neural network—there is no independent, explicitly represented set of beliefs. Third, while the chains of thought chatbots produce look like deliberation, research has demonstrated that the bots often concoct them after the fact, reaching an answer by one route but reporting another.

This is a problem because in the high-stakes applications we all care about, such as medicine, engineering, and scientific research, it matters not only what a system concludes but also how it arrives at its conclusion. When mistakes happen—for example, in medical diagnosis and treatment—we need to be able to pinpoint what went wrong: Was the system’s reasoning at fault, did it draw on invalid evidence, or did it make incorrect assumptions? 

This is why I recently left my position at Google DeepMind. I believe we need a fresh approach to machine reasoning—one that draws on AlphaGo’s architecture. AlphaGo maintains a record of what it knows about a given position: the game tree. This data structure contains all the variations, the possible futures, that AlphaGo has considered, each move and position being annotated with judgments made by its neural networks. As its reasoning progresses, AlphaGo updates the game tree and eventually synthesizes the information in it to decide which move to make. 

Similarly, for general reasoning a system should maintain an epistemic state that represents what the system holds as settled, what it doubts, what it has ruled out, which questions stay open. Reasoning can then be understood as a sequence of moves that change the epistemic state to advance knowledge and reduce uncertainty: deducing consequences, breaking problems into parts, and—crucially—deciding what question to ask, calculation to perform, or experiment to run nex

Acronis
Related Product

Acronis Cyber Backup Cloud Advanced - Server (per Device)

by Acronis

Acronis Cyber Backup Cloud Advanced - Server (per Device) - Enterprise-grade backup & recovery solution

Learn More
t.

Of course, open-world reasoning is harder than playing a board game such as Go or chess. In the real world the current state of affairs is only partially known, the set of available actions is large and variable, and the consequences of actions are stochastic or unknown. 

But recent advances in LLMs and other neural models now give us the capability to take on such general reasoning problems. For example, LLMs can suggest ways of tackling a problem on the basis of what is known and what resources are available. They can interact with tools via APIs or code and help assess whether a claim is supported by available evidence.

Most important, to keep the system honest, an independent part of the system must evaluate each move by how much it actually resolves uncertainty, updating beliefs only when the change is backed by evidence. Once these rules are enforced, the model can accumulate certified knowledge and improve its reasoning policy by learning from past reasoning experiences. You can think of such a system as the scientific method on steroids, with the purpose of producing knowledge that can withstand scrutiny.

I do not think we reach trustworthy machine intelligence by making system 1 bigger. Scale sharpens intuition, but it does not make intuition more deliberative. Move 37 mattered because a machine held a position, weighed the possible futures, and chose the move its artificial instincts would likely have rejected.

Society needs such creative moves in drug discovery, materials, climate, diagnosis—fields where the board looks nothing like a Go board and nobody hands us the rules. We will get such insights only from systems that reason—systems whose conclusions arise from an auditable sequence of evidence, inference, and belief revision rather than from a convincing story told after the fact. 

Thore Graepel is chair of machine learning at University College London. He was a core member of the AlphaGo team at DeepMind and works to ensure that AI benefits human flourishing.

Deep Dive

Artificial intelligence

AI’s recursive self-improvement might not come so quickly after all

AI agents are not yet creative enough to carry out genuinely innovative open-ended AI research, it seems.

Don’t be fooled by this summer of AI hype 

Breathless claims about AGI and new capabilities fall apart pretty quickly under scrutiny.

These startups are chasing the next big thing in LLMs

Meet the new kids nipping at the heels of the AI giants.

Bill Gates says we’ve passed AI’s danger thresholds. Now what?

In a new interview, the billionaire philanthropist sounds an alarm on the urgency of getting our AI policies in order.

Stay connected

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Originally published on October 2, 2026 by MIT.

Expert Analysis for Orange County Businesses

How this impacts SMBs in Southern California

Tech News Roundup: Recent discussions highlight the limitations of large language models (LLMs) in reasoning and creativity, despite their impressive capabilities demonstrated in games like Go. The article emphasizes that while these AI systems can mimic human-like intuition, they lack the genuine reasoning abilities necessary for producing trustworthy and innovative insights.

For small and medium businesses (SMBs) in Orange County and Southern California, understanding the nuances of AI technology is crucial. While leveraging AI can enhance operational efficiency and customer engagement, it is important to remain cautious about its limitations. Relying solely on LLMs without a proper understanding of their reasoning capabilities may lead to misguided decisions or misinterpretations of data. Local businesses should approach AI integration with a balanced perspective, recognizing both its potential and its shortcomings.

At The Network Company, we provide managed IT services and cybersecurity solutions that help SMBs navigate the complexities of technology adoption, including AI. Our AI Services can assist businesses in implementing AI technologies responsibly, ensuring that they complement your strategies without over-relying on their perceived capabilities. Additionally, our cloud backup and BCDR services safeguard your data, enabling you to focus on innovation while maintaining robust protections against potential failures. If you're interested in exploring how we can enhance your technology infrastructure, don't hesitate to contact us.

Read the full article: Original Source Link

Frequently Asked Questions

What makes AlphaGo's reasoning different from LLMs?

AlphaGo combines intuitive guesses with strategic search, while LLMs only predict text tokens.

Why is reasoning important for AI?

Reasoning is crucial for AI to produce trustworthy results and novel insights in complex fields.

Content Quality

Last Reviewed:October 6, 2026
Fact-Checked:October 6, 2026
Sources:6 verified sources
Word Count:1,446 words

Article Tags

Artificial intelligence
App
Opinion
large language models
AI reasoning capabilities
AlphaGo
machine intuition vs reasoning
AI decision-making
cognitive processes in AI
Deep Blue vs AlphaGo
human-like reasoning in AI
artificial intelligence insights
complex decision-making in AI
Go game AI strategies
future of AI technology
T

About Thore Graepel

Content Author

Thore Graepel is a leading expert in artificial intelligence, specializing in machine learning and its applications in strategic decision-making. With a rich background in cognitive science, he explores the intersection of AI and human reasoning.

Expertise: Artificial intelligence, machine learning, cognitive science, strategic decision-making, AI ethics, human-computer interaction, neural networks

Ready to Transform Your Business?

Discover how our expert IT solutions can help your business thrive.