The Scope and Limits of AEO Research, and the Observation Log as a "Breakthrough"

— Beyond Generative Engine Optimization (GEO) —

Date: 2026-01-19 Category: Analysis / AEO / GEO

Reference: Observation Log: Record of AI Structural Referencing Surpassing Human Cognition

0. Introduction: Why AEO Needs Redefinition Now

Currently, methodologies known as AEO (AI Engine Optimization) and GEO (Generative Engine Optimization) are being discussed in both practical and academic fields. However, a decisive "blind spot" exists in these discussions.

That is the fixation on the causal order: "Humans read, and AI summarizes."

In this article, after outlining the status quo and limitations of existing research, we present traffic data observed at GhostDrift Research in January 2026 (see Observation Log). This data demonstrates "AI-side referencing," a phenomenon unexplainable by existing theoretical frameworks, and may serve as a breakthrough in the AEO discussion.

1. The Status Quo of Existing AEO Research

The "optimal solutions" in current AEO/GEO research and practice are established in the following areas. While effective, they remain merely "techniques to be selected by AI."

A. Compliance with Structured Data and Specifications (Schema.org)

Standard methods adopted for Google's "AI Overviews" and Rich Results.

• Method: Implementation of structured data such as FAQPage, HowTo, Article, etc.

• Google's Official Stance: While structured data aids AI understanding, "special optimization" for AI features is not officially required, and creating high-quality content remains the fundamental recommendation

B. Generative Engine Optimization (GEO)

In the academic field, "GEO" is proposed as a method to increase visibility in Search Generative Experiences (SGE).

• Reach of Research: Recent research and practical reports regarding "Generative Engine Optimization (GEO)" suggest that citing authoritative sources, presenting statistical data, and using plain language enhance impressions in generative engines.

• Practical Know-how: The style of placing "Direct Answers" to questions at the beginning of articles has become established.

👉 The Limit So Far

These approaches consistently stand on the perspective of "how to get AI to summarize and cite content made for humans."

2. The "Missing Link" in Current Theory

There are "deficiencies" that existing AEO/GEO theories cannot explain or overlook.

A. Undefined Causal Order

There is an implicit premise that "AI does not value information that humans do not see (or buzz about)." However, the case where "AI comes to reference structural definitions before human cognition" has not been theorized.

B. Inconsistency in Metrics (Traffic vs. Reference)

Existing AEO still places "human clicks (traffic)" as the performance metric. However, with the spread of AI Overviews and the increase in "zero-click searches," a model to measure the value of "Unclicked Reference" is lacking.

C. Conflation of "AI Reference" and "Search Indexing"

In analytics tools like Wix, AI-related Bots (for query response, referencing, and crawling by search generative AI) are measured distinctly from traditional search index bots

. Existing AEO discussions conflate these, missing the perspective of optimizing for "Reference" or "Defining."

3. The Observation Fact as a Breakthrough

The observation log from our institute is an empirical example that breaks through the aforementioned limits. In this paper, we clearly distinguish between (1) Observed Facts, (2) Consistent Interpretations, and (3) Theoretical Implications.

Observed Phenomena

1. AI-Driven Queries: AI-related queries (1,077) overwhelmingly outnumbered human sessions (306).

2. Concentration on Definitions: The reference targets were not flow articles (blog posts) but Definition/Structure pages (Stock) such as "AI Safety" and "Governance."

3. Event-Driven: A distinct spike (surge) was confirmed on January 14th.

While possibilities such as crawler setting changes or increased external links cannot be denied as causes for this spike, even in such cases, the fact of "biased reference to definition pages" remains unchanged and is difficult to explain with existing AEO metrics.

The Meaning of This Data

This is considered not a result of "Optimization by technique," but an "Observation where AI systems came to fixate the Structure (Definition) before human consensus formation."

If existing GEO is a "competition to be selected by AI," this case indicates a state of functioning as an "Infrastructure (Dictionary) for AI."

4. Proposal: Redefinition of AEO

Based on this observation, we propose extending the concept of AEO "upstream" as follows:

Item	Conventional AEO / GEO	Perspective Proposed by GhostDrift
Role	Media "summarized" by AI	Repository providing "Definitions" to AI
Order	Human Evaluation → AI Citation	AI Structure Fixation → Human Reference
Target	Optimization of Answers	Fixation of Structure

As the GhostDrift theory suggests, in a complex society, "structure" is determined before "meaning." Website operators will assume a new role of maintaining "Definition Files" for AI systems, in addition to competing for human attention.

5. Conclusion

Existing AEO research is merely "Chapter One" of Web utilization in the Generative AI era. Our institute's observation log suggests that "traffic absent of humans" is not a failure, but the state of information (structural anchor) in the next generation.

This is a record of a paradigm shift regarding the relationship between the Web and AI, transcending technical theories. This log is preserved as primary source material for a "case of AI-first referencing" in future AEO/GEO research.

The conclusion of this paper is based on observed facts at a specific site and does not claim generalization or determination of causality. It is positioned strictly as a record of phenomena that have not been sufficiently explained by existing AEO theories.

Appendix: Terminology and Data

Terminology Definitions in This Paper

• Flow: Articles and posts consumed chronologically.

• Stock: Pages where definitions, concepts, and structures are described fixedly.

• Definition File: A group of Web pages functioning as a prerequisite structure referenced by AI.

About Data Reproducibility

This log is based on automatic aggregation results by Wix Analytics, and individual queries or time-series values cannot be arbitrarily edited.

Appendix: Summary

This article addresses the lack of discussion on causal order in AEO/GEO research. Based on January 2026 data where AI-driven queries significantly exceeded human sessions, we propose that AI systems may index structural definitions before human consensus forms. This suggests a shift for websites to function as "definition repositories" for AI architectures.

Sources and References

[1]: Google Search Central: "Google Search's AI Overviews and your website". Google official documentation states that special structured data is not required for AI Overviews, and following standard SEO best practices is recommended.

[2]: Recent research and practical reports on GEO (e.g., organization of visibility enhancement methods in generative AI search). Note: This paper refers only to the theoretical positioning and does not evaluate the validity of individual methods.

[3]: Wix Help Center: "Understanding Your Bot Traffic", etc. In Wix analytics, "AI Bots" are explained as a group of bots measured for purposes such as referencing, crawling, and query response for search generative AI.