Thinking about ai visibility
Research, frameworks, and field notes on how language models decide which brands exist.
What Is AIO, and Why It Is Not SEO
SEO optimizes for crawlers that index and rank pages. AIO optimizes for language models that generate answers. Different pipeline, different signals, different strategy.
Read →How AI Systems Actually Decide Which Brands to Recommend
A language model is a function that outputs probability distributions over tokens. Your brand appears when the model assigns it high probability at the right moment in the generation loop.
Read →The Five Levers of AI Visibility
Training corpus, retrieval index, entity graph, real-time grounding, feedback loops. Five levers, each with a different timeline to impact and a different cost to pull.
Read →Why Your Brand Is Invisible to ChatGPT (And What to Do About It)
Three gaps explain most brand invisibility: the pretraining gap, the chunk gap, and the entity gap. Most marketing content never reaches training data, and the reasons are more structural than you think.
Read →Knowledge Graphs, Entities, and How AI Decides Your Brand Exists
There is a difference between being mentioned and being an entity. That difference determines whether AI systems treat your brand as a real thing or a random string.
Read →How We Built a System to Measure AI Visibility (And What We Found)
No one had a way to measure how AI systems see brands. We built a 5-stage pipeline, ran it on 4,000+ brands, and found that 38% are invisible to every model.
Read →