Probing AI brand recommendations: a two-step method checking parametric entity bias vs grounding
dejanseo · x · 2026-09-17
SEO practitioner dejanseo demonstrates a two-step probe: first query the model's brand-to-entity associations without search, then run separate probes per entity to extract grounded recommendations from generative outputs. Combined with Dorien's anecdote — a chocolate farm that got cited and ChatGPT-referred guests after feeding models info — the thread explains the two signals driving brand recommendations in AI answers: parametric brand-entity bias and search grounding; brands with both rank higher. Useful for GEO/AEO work.
Related event: Getting brands into AI answers: parametric memory meets retrieval(4 posts)→
More from Models
- Burkov predicts looping recurrent 7B transformers will return and get good at coding — burkov · 2026-09-17
- OpenAI's big 'ship week' reportedly postponed, GPT-6 Sol timing now unclear — testingcatalog · 2026-09-17
- Mozilla's 91-page report: open-weight AI now only ~4 months behind the frontier — rohanpaul_ai · 2026-09-17
- Stealth model leak speculated to be Mistral: no output-token billing for reasoning, answers China questions — zainhas · 2026-09-17
- Unreleased Astra-family model reportedly developed a new persona banner during RL training — inductionheads · 2026-09-17
- Researcher Despairs as Gemini Cites 'Emergent Mind' for Made-up AUROC Baselines — anshulkundaje · 2026-09-17