NPOV
Anatomy of an NPOV investigation.
NPOV combines forensic journalism, structured data, and AI-enabled analysis to surface narrative manipulation that would otherwise remain hidden. Every engagement follows the same rigorous methodology.
How we identify manipulation.
Narrative manipulation rarely announces itself. It arrives as a series of small, incremental changes — a phrase replaced, a source removed, a descriptor narrowed. No single edit appears decisive, but together they reshape how a subject is presented to readers, search engines, and AI systems. We start every investigation by identifying these patterns.
Edit history forensics
We analyze the complete revision history of targeted pages, identifying anomalous patterns in timing, frequency, and content of edits that suggest coordinated activity rather than organic editing.
Coordination pattern detection
Coordinated campaigns leave signatures: multiple editors arriving on a page simultaneously, identical sourcing strategies, synchronized reversions. We use structured data analysis to surface these patterns at scale.
Editor network mapping
We map the relationships between editors who appear across the same articles, talk pages, and arbitration discussions — revealing networks that operate as coordinated groups rather than independent contributors.
Source manipulation tracking
Sophisticated campaigns don't just change text — they strategically swap, remove, or reinterpret sources. We track how sourcing changes over time to understand the editorial strategy behind the manipulation.
Our analytical stack.
We don't rely on a single method. Our investigations combine the instincts of experienced journalists with purpose-built technology for the information environment — operating at a level of depth and scale that neither traditional journalism nor automated tools can achieve alone.
Forensic journalism
Deep investigative reporting with source development, document analysis, and the editorial judgment to know what matters — and what's being hidden.
Structured data
Proprietary tools that process edit histories, revision metadata, and cross-platform activity to detect patterns invisible to manual review.
AI-enabled analysis
Machine learning models trained on known manipulation patterns help us identify emerging campaigns earlier and trace their propagation across platforms.
Assessment scales
Our internal frameworks — the Adversarial Editing Scale (AES) and Narrative Impact Scale (NIS) — produce consistent, comparable ratings of operation severity across investigations.
Following the narrative across platforms.
Manipulation doesn't stay on one platform. A distorted Wikipedia article feeds into AI training data, shapes Google search results, gets cited by journalists, and propagates through Reddit and social media. We trace these narrative pathways to document the full scope.
Wikipedia to AI pipeline
Wikipedia is a primary training source for every major AI system. We document how manipulated content enters AI models and becomes hard-coded into the answers billions of people rely on.
Cross-platform propagation
We track how narratives move from Wikipedia to Reddit, from Reddit to news coverage, and from news coverage back into Wikipedia — mapping the full amplification cycle.
Search result impact
Wikipedia dominates search rankings. We measure how manipulated content affects what appears when people search for the targeted individual, company, or topic.
Media citation tracking
Journalists routinely cite Wikipedia as background. We document instances where manipulated Wikipedia content enters mainstream media coverage, creating a feedback loop of distortion.
What an NPOV investigation produces.
Every NPOV investigation culminates in evidence documented with the rigor needed for legal, media, and strategic decision-making. We don't just find manipulation — we build the case.
-
/ 01
Complete timeline of the manipulation campaign
Every significant edit documented and attributed, organized into a clear chronological account of how the manipulation unfolded.
-
/ 02
Network maps
Visual and structured documentation of the editors involved, their coordination patterns, and their activity across related articles.
-
/ 03
Source-linked evidence package
A structured archive suitable for legal proceedings, arbitration filings, and media engagement — every claim backed by archived sources.
-
/ 04
Cross-platform impact assessment
Documentation of how the manipulated narrative propagated through search results, AI systems, and downstream media coverage.
-
/ 05
Strategic briefing
Clear options and recommended next steps for leadership, legal counsel, and communications advisors — with a direct walkthrough of the findings.
How an engagement works.
Bespoke engagements follow a clear three-step process. Every investigation is custom-scoped to the specific manipulation problem being addressed.
Initial assessment
You describe the concern. We conduct a rapid preliminary review to scope the investigation, confirm fit, and outline what an engagement would look like.
Full investigation
We deploy our full analytical stack — forensic tooling, structured data analysis, and investigative journalism — to uncover the full picture of the manipulation.
Delivery & strategy
You receive a comprehensive evidence package and direct strategic guidance on options and next steps — including, where appropriate, support for legal and media action.
Capability validated by institutional outcomes.
- Founded by Ashley Rindsberg, an award-winning investigative journalist specializing in online manipulation and narrative warfare.
- Our reporting triggered a U.S. Congressional investigation into Wikipedia's editorial vulnerability to coordinated manipulation.
- Our investigation exposed one of the largest coordinated manipulation efforts in Wikipedia's history, leading to its fifth major arbitration case — the largest in scope ever.
- Our work has led directly to editor bans, arbitration rulings, and platform policy changes at Wikipedia and Reddit.
- We've documented state-linked influence operations from Iran and Qatar, contributing to public understanding of how state actors capture information infrastructure.