Negative Public Relations Research Archive
An open archive of original, independently checkable research on negative PR campaigns, AI reputation attacks, and exploit patterns. Every record carries a stable URL, a concise sourced finding, evidence classification, primary-source links, verification dates, and correction history.
Evidence Classified
Every finding tagged Verified → Unverified
Primary Sources
Direct links to underlying documents
Verification Dates
Each record shows last-verified date
Correction History
All changes logged with editor name
Evidence Classification Methodology
The Negative Public Relations Research Archive classifies every finding on a six-level evidence scale (Verified, Corroborated, Reported, Alleged, Disputed, Unverified) and attributes each finding to the party that produced it — courts, labs, regulators, reporters, or the archive itself.
Google AI Overviews Cited Satirical Reddit Content as Factual (May 2024)
In May 2024, Google's AI Overviews feature produced widely circulated erroneous answers — including a recommendation to use non-toxic glue on pizza cheese — by treating satirical and low-quality Reddit comments as authoritative sources, an incident Google publicly acknowledged.
Coordinated Review Bombing as a Reputation-Attack Vector
Coordinated review bombing — the mass submission of negative reviews by organized groups rather than genuine customers — is a documented reputation-attack vector that platforms including Amazon, Google, Yelp, and Rotten Tomatoes have publicly acknowledged and built detection systems to counter.