The Impossible Test: A 2024 Unsolvable Dataset and A Chance for an AGI Quiz

TL;DR

This paper presents a new evaluation framework using an unsolvable dataset to test large language models' ability to admit ignorance, revealing current limitations and variability across problem types, with implications for AGI development.

Contribution

Introduces a novel dataset and evaluation method for assessing LLMs' uncertainty acknowledgment on unsolvable problems, advancing AGI assessment methodologies.

Findings

Models often fail to admit ignorance on unsolvable problems.

GPT-4 shows higher uncertainty acknowledgment on more difficult problems.

Performance varies significantly across different problem categories.

Abstract

This research introduces a novel evaluation framework designed to assess large language models' (LLMs) ability to acknowledge uncertainty on 675 fundamentally unsolvable problems. Using a curated dataset of graduate-level grand challenge questions with intentionally unknowable answers, we evaluated twelve state-of-the-art LLMs, including both open and closed-source models, on their propensity to admit ignorance rather than generate plausible but incorrect responses. The best models scored in 62-68% accuracy ranges for admitting the problem solution was unknown in fields ranging from biology to philosophy and mathematics. We observed an inverse relationship between problem difficulty and model accuracy, with GPT-4 demonstrating higher rates of uncertainty acknowledgment on more challenging problems (35.8%) compared to simpler ones (20.0%). This pattern indicates that models may be more prone to generate speculative answers when problems appear more tractable. The study also revealed significant variations across problem categories, with models showing difficulty in acknowledging uncertainty in invention and NP-hard problems while performing relatively better on philosophical and psychological challenges. These results contribute to the growing body of research on artificial general intelligence (AGI) assessment by highlighting the importance of uncertainty recognition as a critical component of future machine intelligence evaluation. This impossibility test thus extends previous theoretical frameworks for universal intelligence testing by providing empirical evidence of current limitations in LLMs' ability to recognize their own knowledge boundaries, suggesting new directions for improving model training architectures and evaluation approaches.