Are there ads in ChatGPT and Perplexity?

Ad and answer surfaces are emerging in AI search and evolving quickly. It's early, so the durable move is to make your brand citable (GEO) and keep first-party measurement, rather than betting everything on one new placement.

What is GEO (generative engine optimization)?

Optimizing your content so AI answer engines cite it — clear answer-first writing, structured data, FAQs, an llms.txt, and comparison/glossary pages that directly answer questions.

How do I measure ROI from AI search?

Lean on first-party data: join ad/analytics data with revenue in a warehouse like BigQuery to compute true ROAS, since cross-surface attribution is still immature.

Trends

Advertising in the age of AI search (ChatGPT & Perplexity)

TL;DR

More buyers start with ChatGPT or Perplexity than a search box, and ad and answer surfaces are extending into AI search. It's early and measurement is messy, but the playbook is clear: make your brand citable by AI engines (GEO), keep first-party measurement so you can prove ROI wherever demand originates, and run discovery with reviewable, human-gated AI.

Discovery is moving into AI answers

Instead of a keyword and ten blue links, a growing share of research happens in a conversation with an AI assistant that synthesizes an answer. That changes both where demand forms and how brands get discovered.

GEO: be the source the AI cites

Publish clear, answer-first content with crisp definitions and FAQs that engines can quote.
Use structured data and an llms.txt so AI crawlers can ingest your knowledge base.
Comparison and glossary pages tend to get cited — they answer the exact question.

This is the same generative-engine-optimization thinking behind the shift to agentic systems — see the state of AI in PPC in 2026.

Measurement still needs first-party data

Attribution across AI surfaces is immature, which makes your own data more important, not less. Joining ad and analytics data with revenue in your warehouse lets you measure true ROAS wherever the click came from.

Run it with control

New surfaces mean new ways to overspend on unproven inventory. Keep discovery experiments reviewable and human-gated, and let an agent connect across your tools — including via open standards like MCP — rather than a closed set.

Frequently asked questions

Are there ads in ChatGPT and Perplexity?: Ad and answer surfaces are emerging in AI search and evolving quickly. It's early, so the durable move is to make your brand citable (GEO) and keep first-party measurement, rather than betting everything on one new placement.
What is GEO (generative engine optimization)?: Optimizing your content so AI answer engines cite it — clear answer-first writing, structured data, FAQs, an llms.txt, and comparison/glossary pages that directly answer questions.
How do I measure ROI from AI search?: Lean on first-party data: join ad/analytics data with revenue in a warehouse like BigQuery to compute true ROAS, since cross-surface attribution is still immature.