PwC has allegedly published AI-generated reports containing false or fabricated sources
Quick Answer
PwC's Middle East reports allegedly contain AI-generated content with fabricated sources, particularly the 'Transforming Governance' report, which is 84% likely to be entirely AI-generated.
Quick Take
Despite claims of government reliance on its 'Citizen Pulse' product, no evidence supports these assertions, raising concerns about accuracy and integrity in AI-generated reports.
Key Points
- GPTZero identified false claims in four PwC Middle East reports from 2024-2026.
- The 'Transforming Governance' report is 84% likely to be entirely AI-generated.
- PwC claims governments in Denmark, Saudi Arabia, the U.S., and Australia use 'Citizen Pulse'.
- GPTZero's 'Vibe Citing' highlights loose references without supporting evidence.
- PwC stated it is updating a limited number of citations but didn't explain the errors.
📖 Reader Mode
~1 min readAfter KPMG, PwC appears to be next. GPTZero found fabricated sources and false claims in four PwC Middle East reports (2024–2026). The worst offender, "Transforming Governance," is 84 percent likely to be entirely AI-generated. PwC uses the report to promote a product called "Citizen Pulse," claiming governments in Denmark, Saudi Arabia, the U.S., and Australia rely on it. No evidence supports those claims. Whether this is really an AI problem, though, is a separate question.
Reports with the highest AI scores also had the most false sources. GPTZero calls this "Vibe Citing," where references are cited loosely, often without correct titles, URLs, or authors, and don't actually support the claims they're attached to. The same thing happens in academia.
PwC Middle East told the Financial Times it takes accuracy seriously and is "updating a limited number of supporting citations." The firm didn't explain how the errors got in. GPTZero had already found fabricated sources at KPMG, Deloitte, and Ernst & Young.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
