Academics have raised concerns about the use of ChatGPT in peer reviews of grant applications, prompting the Australian Research Council (ARC) to issue a warning that using generative AI models may breach confidentiality. The ARC_Tracker Twitter account reported instances where assessors’ reports for ARC discovery projects appeared to be generated by ChatGPT, lacking critical analysis or insights. Researchers also found similarities between the AI-generated summaries and their own grant proposals. The ARC has advised peer reviewers against using AI in assessments and plans to update its guidance on AI usage. The use of ChatGPT in peer reviews and ARC’s response to the issue have sparked discussions about workload pressures and the absence of clear policies on AI use.


