NetContentSEO AI Visibility Research & Methodology
NetContentSEO is an independent research and publishing project focused on AI visibility, generative search, retrieval, attribution and the way artificial intelligence systems understand and reconstruct information about brands, authors, concepts and online content.
The project approaches AI visibility as more than a question of whether a brand appears in an AI-generated answer. An entity can be mentioned and still be misunderstood, incorrectly attributed or reconstructed from incomplete information. Our research therefore looks at both visibility and accuracy of representation.
What NetContentSEO Studies
NetContentSEO examines how information moves from published sources into AI-generated answers. This includes how content is retrieved, which sources are selected, how information is attributed and what happens when a model has enough information to recognize an entity but not enough information to describe it reliably.
Current research topics include AI visibility, Generative Engine Optimization (GEO), Retrieval-Augmented Generation (RAG), semantic retrieval, citations, entity recognition, hallucinations and the relationship between traditional search visibility and visibility inside generative systems.
NetContentSEO AI Labs
AI Labs is the experimental section of NetContentSEO. Instead of relying only on assumptions about how large language models behave, we design small reproducible tests and compare the responses produced by different AI systems under the same conditions.
Tests may include commercial AI systems such as ChatGPT, Gemini, Grok and Perplexity as well as smaller models running locally. The purpose is not to produce universal model rankings, but to document observable differences in retrieval, reasoning, attribution, reconstruction and hallucination.
Prompts, test material and model responses are published whenever possible so that readers can understand how a result was obtained rather than seeing only a final score or conclusion.
Explore AI LabsResearch Methodology
NetContentSEO experiments generally begin with a defined question and a controlled input. When multiple AI systems are compared, the same prompt and source material are used wherever the systems allow it. Results are then compared for factual accuracy, consistency, attribution and differences in how the supplied information is interpreted.
Individual experiments are treated as observations rather than scientific benchmarks. A limited number of prompts or model responses cannot establish a universal rule about the behaviour of an AI system. Results are therefore published with their context and limitations and may be repeated as models, retrieval systems and available information change.
This approach is particularly useful for AI visibility research because the question is not simply whether information can be retrieved. We also want to observe whether the relationship between a source, an entity and an idea survives the process that produces the final answer.
The Reconstructability Framework
The Reconstructability Framework is an idea developed through NetContentSEO research to examine AI visibility from the perspective of reconstruction rather than ranking alone.
The central question is whether enough clear, consistent and connected information exists for an AI system to reconstruct an entity or concept correctly when answering a question.
This distinction matters because recognition alone does not guarantee accurate representation. A model may know that a brand, author or idea exists while still reconstructing its attributes, relationships or provenance incorrectly.
What NetContentSEO Is — and Is Not
NetContentSEO is currently an independent research, experimentation and publishing platform. Its published articles, research and AI Labs experiments are freely accessible.
NetContentSEO should not be confused with an automated AI visibility monitoring dashboard. Monitoring platforms primarily measure metrics such as brand mentions, citations and share of voice across repeated prompts. NetContentSEO focuses on researching and testing how AI systems retrieve, interpret, attribute and reconstruct information.
These approaches can complement each other. Monitoring can show whether an entity appears. NetContentSEO's research is primarily concerned with understanding what happens to the information when it does.
Who This Research Is For
The research is intended for SEO professionals, digital marketers, publishers, developers, researchers and organisations trying to understand how generative AI changes online visibility.
It can also be useful to teams investigating how ChatGPT, Gemini, Perplexity, Grok and other AI systems describe their company, products, people or ideas, particularly when the goal is to understand not only whether the entity appears but whether the description is accurate.
Examples of AI Labs Research
Recent experiments have tested whether different AI models can detect false claims hidden inside otherwise accurate technical information, how models review the same deliberately broken PHP application, and how AI systems reconstruct information about an author and the ideas associated with that author.
One experiment found that a small local model attempted to correct a completely fabricated AI industry protocol by generating an alternative history for the same nonexistent protocol. The result provided a useful example of how hallucination can appear during correction rather than only during the initial generation of an answer.
Read the ExperimentAuthorship and Editorial Responsibility
NetContentSEO is independently authored and editorially curated. Research direction, experiments and editorial decisions are overseen by Stefano Galloni, who writes about technical SEO, AI visibility, generative search and the changing relationship between online content and AI systems.
AI tools may be used during experiments or research workflows, but published conclusions are reviewed editorially and experimental results are presented with their limitations rather than treated as universal claims.
Follow the Research
New experiments and observations are published regularly as generative search systems, retrieval methods and AI models evolve.
NetContentSEO AI Labs About NetContentSEONetContentSEO is an independent research and publishing project. Individual AI Labs experiments are observations under specific test conditions and should not be interpreted as universal benchmarks of AI model performance.