Are AI Capabilities Increasing Exponentially? A Competing Hypothesis

2026-02-04

AI Summary

This study challenges the prevailing assumption that AI capabilities grow exponentially. By proposing a two-dimensional decomposition of AI capability—distinguishing between foundational and reasoning capacities—and applying sigmoid (logistic) curve fitting alongside mathematical analysis, we systematically evaluate the sustainability of this growth trajectory. Our findings indicate that the sigmoid fit to current data has already passed its inflection point, and a composite model further supports the conclusion that the overall advancement of AI capabilities may have entered a deceleration phase. These results expose the fragility of mainstream exponential projections and offer a novel theoretical framework for understanding the developmental trajectory of artificial intelligence.

Technology Category

• Humans and AI: Other Foundations of Human Computation & AI
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Rapidly increasing AI capabilities have substantial real-world consequences, ranging from AI safety concerns to labor market consequences. The Model Evaluation&Threat Research (METR) report argues that AI capabilities have exhibited exponential growth since 2019. In this note, we argue that the data does not support exponential growth, even in shorter-term horizons. Whereas the METR study claims that fitting sigmoid/logistic curves results in inflection points far in the future, we fit a sigmoid curve to their current data and find that the inflection point has already passed. In addition, we propose a more complex model that decomposes AI capabilities into base and reasoning capabilities, exhibiting individual rates of improvement. We prove that this model supports our hypothesis that AI capabilities will exhibit an inflection point in the near future. Our goal is not to establish a rigorous forecast of our own, but to highlight the fragility of existing forecasts of exponential growth.

Problem

• AI capabilities
• exponential growth
• forecasting
• inflection point
• sigmoid curve

Innovation

• AI capabilities
• sigmoid growth
• inflection point
• reasoning capabilities
• forecast fragility