JudgeGPT

An empirical platform for evaluating how humans judge AI-generated news on dual axes: source attribution and authenticity.

Main HubLive Demo
RepositorySource code, setup details, and research context.Live DemoTry the platform directly in the browser.Zenodo DatasetJudgeGPT human perception data and related research materials.

Preprints

Industrialized DeceptionCollateral effects of LLM-generated misinformation on digital ecosystems.Eroding the Truth-DefaultHuman susceptibility to foundation model hallucinations and disinformation in the wild.Blessing or Curse?Survey background on generative AI and fake news.

Role in the broader research agenda

JudgeGPT operationalizes the human side of the problem: how people perceive, misclassify, trust, and fatigue under synthetic news conditions.