Five Percent — Video
Five Percent — Graphite
Video

The Future of Search: LLMs, AI Search & Optimization Strategies

36.6K views

Key Takeaways

  • There are 2 goals in AI search: answer optimization (getting your product mentioned in the answer) and citation optimization (getting your page cited as a source).

  • Retrieval Augmented Generation (RAG) improves AI answers by retrieving additional information, often through search, and using it to generate a more accurate response.

  • Large language models predict text one word at a time based on probabilities, which is why asking the same question twice can give you different answers.

  • Co-occurrence matters: the more your product appears alongside relevant terms across the web, the more likely an AI model is to associate it with those topics and mention it in an answer.

  • Reddit and Quora are becoming major citation sources, and they can give completely different results than what shows up in traditional search or on publisher sites.

Summary

Ethan Smith, CEO of Graphite, and Greg Druck, Head of AI at Graphite, explore how to optimize for AI answer engines. The session unpacks the technology behind large language models, including Retrieval Augmented Generation (RAG), and the practical shift from traditional SEO to answer optimization.

Ethan and Greg cover answer and citation optimization, and how user-generated content and co-occurrence patterns influence what AI cites. They also discuss the role of owned versus earned media, how to track visibility in LLM answers, and how to think about ads versus organic in an AI-first search landscape.

The discussion spans ChatGPT, Perplexity, and other AI surfaces, with a Q&A section covering practical tools, SEO playbooks, and where the discipline is headed next. The takeaway: AI doesn't replace SEO, it expands it, and the brands that win will be the ones that adapt their content strategy to the new mechanics of answer engines.

The narrative right now is that large language models are going to make traditional search obsolete, but actually answer engines are building on top of traditional search engines.

Greg Druck, Head of AI at Graphite