GPTZero Flags Fabricated Citations in PwC Middle East Reports
The findings come amid broader scrutiny of AI-assisted research after similar issues emerged at KPMG, EY, and Deloitte.
News
- Emirates NBD Partners With Dubai Future District Fund for Fintech Innovation
- UAE Steps Up Agentic AI Push Across Federal Government
- Mubadala Eyes Up to $6.3 Bn Investment in Japan AI Data Center
- Alibaba Plans Revenue-Sharing Model for Next Open-Weight AI Release
- AI Dispatch: Global Reorder and a Region on the Rise
- Ajman Targets 100 AI Initiatives Under Government-Wide Program
[Image: Nomita Samaiyar/MITSMR Middle East]
Consulting giant PwC is facing scrutiny after an investigation by AI detection company GPTZero found that four thought leadership reports published by PwC Middle East between 2024 and 2026 contained fabricated citations, unsupported claims, and other apparent AI-generated research errors.
The analysis identified recurring issues across the reports, including hallucinated references, unverifiable statistics, misattributed sources, and claims that could not be substantiated through publicly available records. One report, Transforming Governance (2025), was assessed by GPTZero as having an 84% probability of being AI-generated, rising to 100% when its reference section was excluded.
The report introduced PwC’s “Citizen Pulse” governance framework and claimed it had been implemented by governments in Denmark, Saudi Arabia, the United States, and Australia. GPTZero said it found no publicly available evidence supporting those claims.
Researchers Paul Esau, Om Ogale, and Alex Cui also questioned the report’s citation practices. Of the 17 references listed, only 10 were linked to footnotes in the body of the document, a structure they said departs from standard research conventions and makes independent verification more difficult.
According to GPTZero, the remaining three reports exhibited similar issues, including the conflation of unrelated datasets, citations that did not support the claims they accompanied, and causal relationships presented without supporting evidence. One cybersecurity report, Building a Cyber-Resilient eMobility Ecosystem, included a citation URL referencing ChatGPT rather than a primary source.
PwC told the Financial Times that it “takes the accuracy of our published research seriously” and is updating “a limited number of supporting citations” in the identified reports. The firm added that it has quality control processes for research and content development as part of its responsible AI practices.
The findings come amid broader scrutiny of AI-assisted research across the consulting industry. GPTZero has reported similar issues in a KPMG paper on agentic AI, while EY withdrew a consumer loyalty study earlier this year after discovering fabricated footnotes and apparent AI-generated content. Deloitte also partially reimbursed an Australian government contract in 2025 after AI-related errors were identified in a commissioned report.
