James Dooley and Dan Petravvic discuss how AI SEO differs from traditional SEO because large language models rely on probabilistic selection, brand familiarity and internal training bias rather than simple keyword matching. Dan explains that optimising for Gemini, ChatGPT, Claude and Perplexity requires more than rankings since AI assistants select brands based on confidence scores and entity recognition. They explore selection rate optimisation, model bias, grounding citations and how Treewalker.ai surfaces low confidence tokens to strengthen brand positioning. The conversation highlights why branded search, user engagement signals and knowledge graph presence increase AI visibility because models prefer familiar, authoritative entities over thin exact match domains. They also cover generative interfaces, agentic AI, UserLM simulations and how synthetic user sessions help test AI selection behaviour.Where to Listen to This EpisodeOptimising AI Search Visibility is available across all major podcast platforms. Choose your preferred platform below.Listen to Optimising AI Search Visibility on TransistorListen to Optimising AI Search Visibility on Pocket CastsListen to Optimising AI Search Visibility on Amazon MusicListen to Optimising AI Search Visibility on AudibleListen to Optimising AI Search Visibility on CastroListen to Optimising AI Search Visibility on CastboxListen to Optimising AI Search Visibility on Podcast AddictListen to Optimising AI Search Visibility on DeezerListen to Optimising AI Search Visibility on RephonicListen to Optimising AI Search Visibility on MetacastListen to Optimising AI Search Visibility on Player FMListen to Optimising AI Search Visibility on SpotifyListen to Optimising AI Search Visibility on Listen NotesListen to Optimising AI Search Visibility on Podcast GuruListen to Optimising AI Search Visibility on PodchaserListen to Optimising AI Search Visibility on Podverse