AI Search Visibility in 2025: What's New and What's Changed
The state of AI search in 2025 — what's changed, what the data shows, and what enterprise SEO teams are doing about it.
AI Search Has Gone Mainstream
In 2025, AI-powered search is no longer experimental. Google AI Overviews appear in over 50% of informational queries in the United States, fundamentally changing how content is surfaced for question-type searches. Traditional position-1 organic rankings still matter for navigational and transactional queries, but for informational intent — how to, what is, best for — AI Overviews are increasingly the first answer a user sees.
ChatGPT Browse Is a Search Engine
ChatGPT Browse capability is now processing millions of queries daily, functioning as a de facto search engine for a generation of users who prefer conversational interfaces over link lists. ChatGPT citation patterns favor pages with strong brand mentions in trusted external sources — Reddit, Wikipedia, G2, YouTube, and established media — over pages that merely optimize their own content. Getting into ChatGPT responses requires being talked about, not just being technically optimized.
Perplexity Growth and Citation Model
Perplexity has grown to over 15 million daily active users in 2025 and uses a hybrid retrieval model that explicitly shows citations to users. This makes Perplexity the most transparent of the major AI search engines about how it sources answers. Perplexity prioritizes recent, cited content — articles published in the last 6 months with external citations perform significantly better than older content. If you are creating content to appear in Perplexity, recency and source attribution within your own articles are key signals.
The Rise of AEO as a Discipline
Enterprise SEO teams at Fortune 500 companies began hiring dedicated AEO specialists in 2024, and the role has expanded rapidly in 2025. The skill set is distinct from traditional SEO: AEO specialists focus on entity management, source graph building, structured data implementation, and AI attribution tracking — not keyword research and link building. Digital marketing agencies are adding AEO services to their offerings, and tools like AEO Track AI represent the emerging category of AI visibility measurement platforms.
The Ghost Citation Problem
61.7% of AI citations are ghost citations — the AI uses the URL as a source when formulating its answer, but the brand name never appears in the visible response. This means brands are influencing AI answers and getting zero credit for it. Ghost citation risk is highest for brands with weak entity clarity and brands in competitive categories where AI engines blend multiple sources without naming any of them. Solving ghost citations requires building stronger brand entity signals, not just better content.
What the Data Shows by Industry
B2B SaaS pages with FAQ schema and direct-answer openers are being cited 3x more frequently than pages without those features. Healthcare, finance, and legal content is seeing the fastest growth in AI citation rates — these high-trust, expert-content industries align well with what AI engines prioritize. E-commerce pages are the most affected by ghost citations, where product information is extracted by AI engines but the specific retailer is rarely named.
What to Do Right Now
The teams making the most progress in 2025 are doing three things: auditing their top 20 pages for direct-answer structure and fixing the openers, adding FAQ schema to every page that answers a common buyer question, and building their source graph by getting the brand mentioned in Reddit, G2, LinkedIn, and Wikipedia. These are the high-ROI moves based on current citation data.