arXiv:2606.05383v1 Announce Type: cross Abstract: Can artificial intelligence (AI) refute economic theory? I document experiments in which I asked several AI models (Gemini, Refine, Claude, and ChatGPT) to check the correctness of four published papers in economic theory, each containing an error that I helped identify or correct. ChatGPT Pro performed best, occasionally constructing counterexamples and corrected proofs, while other models fared worse. However, no model located a true error without substantial human guidance, and data contamination complicates interpretation. I argue that a co
The rapid advancement and accessibility of large language models are pushing researchers to test their capabilities in complex analytical tasks, specifically within academic fields like economics.
This research provides a nuanced understanding of current AI model limitations in critical thinking and complex problem-solving, tempering expectations while also highlighting areas for human-AI collaboration.
The perceived infallibility or complete autonomy of AI in complex analytical work is moderated, emphasizing that human expertise remains crucial for validating and guiding AI outputs.
- · AI developers focused on advanced reasoning
- · Economists augmenting research with AI tools
- · Human domain experts
- · Those expecting fully autonomous AI academic agents
- · AI models lacking robust reasoning capabilities
AI tools will increasingly be used as assistants in academic research, but always under human supervision.
The development of AI models specifically designed for logical deduction and error identification in complex systems will accelerate.
The definition of 'original research' may evolve to include AI-assisted discovery, raising questions about intellectual property and attribution.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI