AI Gamestore: Scalable, Open-Ended Evaluation of Machine General Intelligence with Human Games
🤖 AI Summary
Current AI evaluation benchmarks are often confined to narrow tasks, prone to overfitting, and inadequate for assessing general intelligence. This work proposes the “Human Game Multiverse” framework, which conceptualizes the space of all conceivable human games as an open-ended evaluation domain. To operationalize this vision, the authors introduce AI GameStore—a platform that leverages large language models and human-in-the-loop mechanisms to automatically extract, generate, and containerize executable game variants from sources such as the App Store and Steam. This enables dynamic, scalable assessment of general intelligence. The project has produced 100 representative games; evaluations on seven state-of-the-art vision-language models reveal that even the best-performing model achieves an average score below 10% of human performance, exhibiting pronounced deficiencies in world modeling, memory, and planning capabilities.
Application Category
📝 Abstract
Rigorously evaluating machine intelligence against the broad spectrum of human general intelligence has become increasingly important and challenging in this era of rapid technological advance. Conventional AI benchmarks typically assess only narrow capabilities in a limited range of human activity. Most are also static, quickly saturating as developers explicitly or implicitly optimize for them. We propose that a more promising way to evaluate human-like general intelligence in AI systems is through a particularly strong form of general game playing: studying how and how well they play and learn to play \textbf{all conceivable human games}, in comparison to human players with the same level of experience, time, or other resources. We define a "human game" to be a game designed by humans for humans, and argue for the evaluative suitability of this space of all such games people can imagine and enjoy -- the "Multiverse of Human Games". Taking a first step towards this vision, we introduce the AI GameStore, a scalable and open-ended platform that uses LLMs with humans-in-the-loop to synthesize new representative human games, by automatically sourcing and adapting standardized and containerized variants of game environments from popular human digital gaming platforms. As a proof of concept, we generated 100 such games based on the top charts of Apple App Store and Steam, and evaluated seven frontier vision-language models (VLMs) on short episodes of play. The best models achieved less than 10\% of the human average score on the majority of the games, and especially struggled with games that challenge world-model learning, memory and planning. We conclude with a set of next steps for building out the AI GameStore as a practical way to measure and drive progress toward human-like general intelligence in machines.
Problem
Research questions and friction points this paper is trying to address.
general intelligence evaluation
human games
AI benchmarking
open-ended assessment
machine intelligence
Innovation
Methods, ideas, or system contributions that make the work stand out.
general game playing
human games
open-ended evaluation
AI GameStore
vision-language models
🔎 Similar Papers
No similar papers found.
