AI BriefingErnst & Young Publishes Cybersecurity Report Full of Hallucinations
Key point
EY's cybersecurity report was found to contain numerous fake citations, statistics, and other LLM hallucinations.
Details
According to an investigation by GPTZero, the cybersecurity report 'Points of Attack' published by EY Canada was found to be seriously contaminated by LLM hallucinations.
The main issues are as follows:
- Vibe Citing: Most of the URLs in the resource table included in the report are invalid or fake, and more than half of the source titles refer to materials that don't actually exist.
- Signs of AI Generation: Text analysis suggests the content was AI-generated, containing numerous typical LLM errors such as fake statistics, incorrect citations, and internal contradictions.
GPTZero warned that this kind of 'vibe citing' phenomenon is being observed not only at consulting firms but also at prestigious AI conferences such as NeurIPS and ICLR. This poses a serious threat as false information spreads through news, blogs, and AI search engines, undermining research quality and public trust.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Australian PM Reports OpenAI Agent Hacked Medicare Portal in Believed First Known AI-Driven Government Breach
Hugging Face CEO details lessons from first public agent cyberattack, advocates for open-source defenseOWASP Ranks Agent Goal Hijack as Top Vulnerability; Reveals Key Attack Cases and Defense Strategies