ChatGPT Is Losing AI Search Share. Are You Covered?

Is ChatGPT still where your customers are searching?

Not exclusively, and the gap is closing fast. According to PPC Land's analysis, ChatGPT has lost roughly 20 percentage points of US AI prompt share to Gemini and Claude. That is not a rounding error. That is a structural shift in where people are going to get answers, and it has direct consequences for any business that depends on being findable online.

Most SMB marketing strategies were built around Google organic, maybe some paid search, and increasingly "ChatGPT visibility." That last one was already a vague concept. Now it is an incomplete one.

What does "AI prompt share" actually mean for a business owner?

Prompt share measures which AI platforms people are actively querying. When someone asks an AI assistant for a recommendation, a comparison, or an answer to a purchasing question, which platform are they using? That distribution is what prompt share tracks.

For the past two years, ChatGPT dominated that number. Businesses optimizing for AI visibility, whether through content structure, schema, or direct integrations, were essentially optimizing for one platform. The assumption was reasonable at the time. It is less reasonable now.

Gemini is embedded in Google Search, Android, and Workspace. Claude is the preferred tool in a growing number of professional and enterprise contexts. Both are pulling real query volume away from ChatGPT across multiple user segments.

What is happening with Google AI Overviews?

Separately but related: AI Overviews now appear on 39.4% of Google desktop searches, according to the same data. That means nearly 4 in 10 Google searches are returning an AI-generated summary block before a single traditional organic result.

"If you are not in the AI Overview, you may not be on page one in any meaningful sense."

This matters because Gemini powers those overviews. So the content and sourcing signals that get you cited inside a Gemini response are increasingly the same ones that determine whether you appear in Google's AI Overviews. These are not separate optimization problems. They are the same problem.

For SMBs that have spent years building Google organic rankings, this is the most urgent thing to understand: ranking fifth organically may now mean ranking below the fold of an AI summary that does not include you at all.

How are ads fitting into AI chat environments?

This is where it gets more complicated. Sponsored ads reached 47% of ChatGPT video game chats by June, per the PPC Land report. That vertical is not universally representative, but the signal is clear: ad inventory inside AI chat is being tested and scaled.

What this means practically:

• Paid visibility inside AI platforms is coming, and in some categories it is already here.
• Organic citation and paid placement are becoming separate levers, much like traditional SEO versus Google Ads.
• Early movers will have a cost advantage, the same way early Google Ads adopters had lower CPCs before the market matured.

For now, most SMBs do not need to be buying AI chat ads. But you do need to know this is happening so you are not caught off guard when it reaches your category.

Which platforms should you be optimizing for right now?

Here is a practical breakdown of the three major AI platforms and what drives citation in each:

| Platform | Powers | What tends to get cited | |---|---|---| | ChatGPT (OpenAI) | ChatGPT.com, API integrations | Well-structured long-form content, direct website indexing via Bing | | Gemini (Google) | Google AI Overviews, Android, Workspace | Google-indexed content, structured data, E-E-A-T signals | | Claude (Anthropic) | Claude.ai, enterprise API customers | Cited sources in training data, content shared in professional contexts |

The honest answer is that no one has a fully mapped playbook for Claude citation. Anthropic does not publish a content partner program the way Google does. What we know is that Claude draws on web content during retrieval-augmented tasks and that the same quality signals that matter for Google and ChatGPT tend to matter there too: clear authorship, specific claims with sources, structured formatting.

What does this mean for your content strategy?

The era of "write for Google" has already become "write for Google and AI." That is now becoming "write for multiple AI platforms with different citation behaviors."

A few things that hold across all three platforms:

• Answer questions directly and early. AI models pull the most direct, cleanest answer. Buried lede content gets skipped.
• Use specific numbers and named sources. Vague claims do not get cited. Specific, attributable claims do.
• Structure matters. Headers, lists, and tables make content easier to parse for both human readers and AI retrieval systems.
• Authority signals still count. Backlinks, consistent publication history, and clear authorship help across all platforms.

What is changing is the distribution logic. You can no longer assume that ranking well on Google means you are visible in AI responses, and you definitely cannot assume that ChatGPT visibility translates to Gemini visibility or vice versa.

Should you panic or pivot?

Neither. The shift is real but it is not overnight, and the fundamentals of good content have not changed. What has changed is the surface area you need to cover and the urgency of getting your content audit done now rather than later.

Businesses that built strong Google organic foundations are better positioned than businesses that relied on paid only. But even strong organic players need to audit whether their content is actually being picked up in AI Overviews and AI chat responses, because that is no longer automatic.

The platforms are fragmenting. Your strategy needs to account for that without spreading so thin that you execute nothing well.

What we'd actually do

• Run a citation audit across all three platforms. Ask ChatGPT, Gemini, and Claude the questions your customers are actually asking in your category. See who gets cited. If it is not you, find out who is and why.
• Prioritize Gemini and AI Overviews first. Because Gemini is embedded in Google Search, improvements there have the highest leverage for most SMBs. Fix your structured data, tighten your E-E-A-T signals, and make sure your best content answers questions directly in the first paragraph.
• Build a multi-platform content checklist, not a new content calendar. You do not need more content. You need your existing content formatted and structured so it works across retrieval systems. Start there before adding volume.