Problem
Research questions and friction points this paper is trying to address.
Artificial General Intelligence
Complex Systems
Benchmarks
Church-Turing-Deutsch Principle
Non-ergodic Environments
Why machines will still not rule the world
🤖 AI Summary
This study systematically challenges the prevailing optimism regarding the feasibility of Artificial General Intelligence (AGI). By integrating complexity science, thermodynamic principles, and a physicalist perspective, it presents an empirical critique of existing machine learning benchmarks. The findings demonstrate that current models fail to capture the emergent behaviors inherent in complex systems, that improvements in benchmark scores are largely spurious artifacts of data contamination, and that cognitive models exhibit fundamental limitations within open, non-ergodic environments. Furthermore, this work provides a mathematical argument for the infeasibility of AGI, delineating the theoretical boundaries of machine intelligence. Ultimately, it establishes a novel paradigm for the objective evaluation of AI capabilities, cautioning against overestimating the potential of contemporary approaches to achieve general intelligence.
📝 Abstract
In our book Why machines will never rule the world [13, 14] we argue that arti- ficial general intelligence is mathematically impossible. This is because the human beings and the processes which exhibit intelligence are complex systems whose be- haviour cannot be captured by the kinds of models that we can generate with or without computers. Proponents of contemporary machine intelligence respond with two lines of argument: a theoretical one, grounded in the universal approximation theorems for neural networks and the Church-Turing-Deutsch principle; and an em- pirical one, grounded in rapidly rising scores on standardized benchmarks. In this communication we examine and reject both responses. First, we show serious issues in the physicalist counter-argument based on the Church-Turing-Deutsch principle. Second, we review recent evidence to the effect that prominent benchmarks are compromised by training-data contamination, flawed test construction, and strate- gic optimization. Our central argument remains: That models required to perform cognitive behaviour in open-ended, thermodynamically complex and non-ergodic environments are not and will not become achievable.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Artificial General Intelligence
Church-Turing-Deutsch Principle
Benchmark Contamination
Non-ergodic Environments
Complex Systems
🔎 Similar Papers
