Over the Edge of Chaos? Excess Complexity as a Roadblock to Artificial General Intelligence
Not on the map. Why?
Abstract (arXiv)
In this study, we explored the progression trajectories of artificial intelligence (AI) systems through the lens of complexity theory. We challenged the conventional linear and exponential projections of AI advancement toward Artificial General Intelligence (AGI) underpinned by transformer-based architectures, and posited the existence of critical points, akin to phase transitions in complex systems, where AI performance might plateau or regress into instability upon exceeding a critical complexity threshold. We employed agent-based modelling (ABM) to simulate hypothetical scenarios of AI systems' evolution under specific assumptions, using benchmark performance as a proxy for capability and complexity. Our simulations demonstrated how increasing the complexity of the AI system could exceed an upper criticality threshold, leading to unpredictable performance behaviours. Additionally, we developed a practical methodology for detecting these critical thresholds using simulation data and stochastic gradient descent to fine-tune detection thresholds. This research offers a novel perspective on AI advancement that has a particular relevance to Large Language Models (LLMs), emphasising the need for a tempered approach to extrapolating AI's growth potential and underscoring the importance of developing more robust and comprehensive AI performance benchmarks.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 25 of its 73 references that are in the map. First posted on arXiv 4 Jul 2024.
Builds on
25 of 73 references.
References · 25
- 2017Attention Is All You Needcited by 60,361
- 2023GAIA: a benchmark for General AI Assistantscited by 1,128
- 2022Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasetscited by 629
- 2023Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervisioncited by 467
- 2022Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Taskcited by 431
- 2023Augmented Language Models: a Surveycited by 403
- 2022DayDreamer: World Models for Physical Robot Learningcited by 400
- 2023Evaluating Large Language Models in Theory of Mind Taskscited by 222
- 2020Liquid Time-constant Networkscited by 161
- 2023The Generative AI Paradox: "What It Can Create, It May Not Understand"cited by 98
- 2024Can Large Language Models Reason and Plan?cited by 82
- 2024No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performancecited by 75
13 more
- 2024Inadequacies of Large Language Model Benchmarks in the Era of Generative Artificial Intelligencecited by 68
- 2024Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Modelscited by 68
- 2023When Brain-inspired AI Meets AGIcited by 54
- 2023Grokking as a First Order Phase Transition in Two Layer Networkscited by 48
- 2023Brain-inspired learning in artificial neural networks: a reviewcited by 26
- 2024Human-like object concept representations emerge naturally in multimodal large language modelscited by 25
- 2022AI Autonomy : Self-Initiated Open-World Continual Learning and Adaptationcited by 23
- 2023From Google Gemini to OpenAI Q* (Q-Star): A Survey of Reshaping the Generative Artificial Intelligence (AI) Research Landscapecited by 18
- 2024The Generative AI Paradox on Evaluation: What It Can Solve, It May Not Evaluatecited by 16
- 2016Superintelligence cannot be contained: Lessons from Computability Theorycited by 13
- 2019Optimal Machine Intelligence at the Edge of Chaoscited by 7
- 2021Self-Organized Criticality in the Braincited by 7
- 2021Optimal input representation in neural systems at the edge of chaoscited by 2
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.