Why LLMs and Other AI Systems Cannot Reason
Jon Salisbury
CEO @ Nexigen - Ultra Curious, Humble - Cyber Security, Cloud, Smart City, AI, Quantum Physics, Psychology, Leadership
Understanding the Limitations of LLMs
Intro:
It's important to clarify that the purpose of this discussion is not to discourage the use of AI but to ensure a clear understanding of its capabilities and limitations. Using AI effectively requires expert guidance, as proper implementation is complex and nuanced. This article delves into why large language models (LLMs) and similar AI systems do not possess true reasoning abilities.
Anthropomorphism and AI
One common pitfall is the anthropomorphism of AI—attributing human traits and abilities to machines. This is a misconception, especially when it comes to reasoning and intelligence.
Testing AI Reasoning
Principle of the Test:
The Experiment:
A test was conducted using GPT-4 and Claude. They were asked to create and describe a model of a normal-sized home, answering detailed questions about it.
Prompt:
"You are an intelligent robot who can build a world model of something in memory. I want you to place a model of a normal-sized home in your mind now. You will want to ensure you go over this home in detail and be prepared for questions about the entire home inside and out."
Response:
The AI generated a detailed description of the home, including its exterior and interior, down to specific room details and dimensions.
Follow-up Questions and Responses:
领英推荐
Inconsistencies and Analysis:
When asked to provide specific room sizes and reconcile the total square footage, discrepancies arose. The AI's responses showed inconsistencies, such as:
Key Findings:
The Nature of Reasoning
What is Reasoning?
Reasoning involves more than just generating plausible responses. It requires:
Advanced Reasoning Concepts:
Researchers like Yann LeCun are exploring advanced AI solutions, such as Visual-Jeffersonian Epsilon Probability Algorithm (VJEPA), to enhance AI's reasoning capabilities. However, current LLMs like GPT-4 fall short of true reasoning.
My Opinion:
While LLMs are powerful tools for generating text, guess the correct pixel and can appear to understand and reason, they fundamentally operate on pattern recognition and token prediction. True reasoning and "knowing" remain beyond their capabilities, requiring more advanced AI models and approaches. Understanding these limitations is crucial for their effective and responsible use in various applications.
Remember language to them is just the frequency and relationships of morse code streams that come over via electrons:)
For more information please follow me on my daily thoughts and exploration of life... This article was written with ai assistance. Open AI, Perplexity and a few other engines for analysis and testing.
Accepting vCISO Clients for 2025 | Helping SMBs Grow by Enabling Business-Driven Cybersecurity | Fractional vCISO & Cyber Advisory Services | Empowering Secure Growth Through Risk Management
3 个月Insightful Jon Salisbury
VP of Marketing at TechUnity, Inc.
4 个月From an industry perspective, this article clarifies LLMs' current limitations, essential for realistic AI application development.
Founder @ InnovationAcceleration.ai | Innovation, Artificial Intelligence
4 个月Love the conversation, but I'd tweak the question. Do LLMs demonstrate human reasoning? The short technical answer is no, the complex answer is to some degree, but like you highlight in the article it's different. Can the current combination of LLMs and human reasoning outperform human minds alone? Look at Chess after Watson, or Go after AlphaGo. Top players now regularly use AI models to analyze positions and variations at superhuman depths, uncovering insights previously inaccessible to human minds alone. LLMs are showing the same capabilites to augment human reasoning. We are all wrong. ??
Founder | Public Policy| AI| Blockchain| Ecosystem Builder| Keynote Speaker.
4 个月I’ve said it a few times artificial intelligence is neither artificial nor intelligent. I agree with this assessment.
Director of Startup Services ● Ecosystem Enabler ● Growth Catalyst ● Investor Relations ● Incubation and Innovation Leader
4 个月While different from human reasoning, ai offers limited (yet expanding) simulated reasoning. It has definitely been helpful to me in answering some of life's biggest questions and may offer insights into the nature of knowledge and reality itself! An adjacent conversation I had with claude about simulated perspective attached...