Is GEO Dead? I Checked Every Study Used to Bury It

Summary

Guerin Green published a video and a companion site that audit an article by Shaun Anderson of Hobo Web. The article argues that generative engine optimization has been disproven. Green went back to each study the article cites. He agrees that most GEO tactics do not hold up and that the 2023 paper which named the field had real flaws. He also finds that one quoted sentence does not appear in the study it is attributed to, and that a lab experiment is presented as if it described the live web.

The audit points out that the Toronto study the article relies on found 97% of the pages ChatGPT retrieved were outside Google's top 50. That finding cuts against the article's own argument. The companion site places each claim next to the primary source, covers the five practitioners the article names, and logs Green's own corrections, including one dated 23 October 2026 about how the Toronto paper's authors were listed.

Demonstration by Guerin Green, embedded from YouTube. AIOTruth did not reproduce the test.

Why it matters

Brands deciding whether to invest in AI visibility hear two loud positions: vendors who sell GEO tactics and critics who say the field is dead. This audit shows that the research supports neither extreme. Many content tricks fail when tested, but ChatGPT's retrieval draws heavily on pages outside Google's top results. That means ranking well in classic search does not settle whether a brand gets cited in AI answers. Anyone trying to be found or recommended by AI systems gets a practical method here: check each cited study against what it actually says before acting on either side's conclusion.

Source

• geocoffin.novcog.us.com/ the item itself, published by Guerin Green, created by Guerin Green
• arxiv.org/abs/2602.12187 Original dataset with disclosed methods
• arxiv.org/abs/2610.15033 Original dataset with disclosed methods
• arxiv.org/abs/2311.09735 Original dataset with disclosed methods
• geocoffin.novcog.us.com/the-misquote Named expert analysis, or a first-party claim about itself, first party
• youtube.com/watch?v=6xIpBRgqaWcwhere AIOTruth found it

What was checked

• canonical URL taken from the page itself
• read the item page (200)
• https://arxiv.org/abs/2506.11097: does not mention the item, tier 5 (Original dataset with disclosed methods)
• https://arxiv.org/abs/2602.12187: supports the item, tier 5 (Original dataset with disclosed methods)
• https://arxiv.org/abs/2610.15033: supports the item, tier 5 (Original dataset with disclosed methods)
• https://arxiv.org/abs/2311.09735: supports the item, tier 5 (Original dataset with disclosed methods)
• https://geocoffin.novcog.us.com/: supports the item, tier 8 (Named expert analysis, or a first-party claim about itself)
• https://geocoffin.novcog.us.com/the-misquote: supports the item, tier 8 (Named expert analysis, or a first-party claim about itself)

Limitations

• The demonstration was not independently reproduced by AIOTruth.

AIOTruth judgment

The item is rated Supported because its central points can be traced to primary sources: the original GEO paper and two later arXiv studies with disclosed methods are among the confirmed sources. Relevance scores 9 because the question of whether GEO works sits at the center of AI-era discovery. Evidence also scores 9 because claims and source quotes appear side by side and the author logs his own errors publicly. Usefulness scores 6 because the work audits someone else's argument and gives limited direct guidance on what to do. Originality scores 5 because it is a response piece, not new research. AIOTruth did not reproduce the analysis, so readers should consult the cited papers directly.

Read the source Open the source

Found something wrong here? AIOTruth corrects material errors openly. Have something we should review? Submit a find.