You've seen the slide. The deck makes a GEO claim, a footnote sits under it, and the journal's name sounds like proof.
A 2026 review in MDPI's Electronics now gives one GEO paper that treatment in print. It lists a "GEO-First" framework among five prior frameworks and calls it "particularly significant." SEO consultant Shaun Anderson says the journal that published the paper is a cheap route to a scholarly DOI.
My read is that a citation tells you someone cited the paper. It does not tell you anyone tested the method. Check the journal before you trust the DOI.
Anderson laid out his case in an X article on October 16, 2026.
He says he found the paper while auditing how scholarly indexes pick up his own AI SEO writing. The paper, A GEO-First Framework, ran in the World Journal of Advanced Research and Reviews. The journal's page shows it was received on December 11, 2025, and accepted on January 29, 2026, about seven weeks later.
Anderson has a stake, and he says so. He argues the paper's central concepts, including inference optimisation and the Fortress of Facts, come from his 2025 work, and that the bibliography cites his site without his name. He calls the whole thing "Parasite SEO for the Mentions Economy, dressed up in a lab coat." That is his framing, not mine. He also points out that the author runs an online reputation management agency, and he left the author's name out of his post. I am doing the same.
His case against the journal is that it is cheap and fast. The journal's own page advertises an impact factor of 8.2. Anderson says that number comes from vanity directories, not Clarivate. He puts the processing fee at about $40 and says the paper got a Crossref DOI that Google Scholar and other indexes ingest on their own.
In its own text, the paper says direct empirical studies of the AI-mention-to-branded-search pathway "remain limited," and that the framework "has not been validated through controlled experimentation." It draws on practitioner publications for ground it calls "too recent for peer-reviewed coverage." Anderson also flags a table that says Interaction to Next Paint "Measures the responsiveness of a page for agentic scraping." He notes that INP is a Core Web Vital for how fast a page answers a human tap or click.
The paper has already been cited. The Electronics review was published on August 10, 2026, by Konstantinos Roumeliotis and colleagues. It lists the paper as reference 6 and puts its framework in a comparison table with four others.
The review's authors call the framework "particularly significant" because it moves the question from how to rank to how to become cited. They also credit it with introducing Share of Model as a measure of how often an AI system references a brand. They then propose a framework of their own.
A review is supposed to summarize what exists, and that alone is no scandal. But citing a framework is not the same as testing it, and the paper itself says it has not been tested. That is the gap Anderson is pointing at.
Disclosure: my agency, ViewEngine, sells AI visibility work and builds its own tracking tools.
Anderson calls this the mentions economy. If independent mentions build the consensus AI answers lean on, a scholarly citation could carry extra weight, whether or not the research holds up. That is my hypothesis, and nothing in these sources shows an AI system citing this paper. My guess is the paper will get cited far more often than its method gets tested. When a GEO claim arrives with a journal behind it, look up the journal, the review timeline and who stands to profit before you pass it on.
Free newsletter
The Sherpa Brief
One email a week on what changed in AI search, what it means, and what to do about it. Written for marketers, SEOs, and publishers.
Free. Unsubscribe in one click.



