AI Overviews Are Here: How Expert Content Gets Cited

Type a question into Google today and you may not see ten blue links first. You may see a paragraph.

That paragraph is an AI Overview, and it has been showing up for US searchers since May. It changes the question every expert with a website should be asking. Not "how do I rank?" but "how do I become one of the sources that paragraph is built from?"

In fifteen years of working for myself, I have watched search change shape more than once. Here is my honest read on what is known, what is not, and how I would write for it. I will mark the difference as I go.

What changed when AI Overviews launched?

On May 14, 2024, Google announced it was bringing AI Overviews to everyone in the US, and said it expected to reach over a billion people by the end of the year.¹ The features run on a Gemini model customized for Search.¹

Where the links went

Google says the links inside AI Overviews get more clicks than a traditional listing for the same query would, and that ads continue to appear in dedicated slots with clear labels.¹ That is Google's claim, and I would treat it as a claim until independent data shows it for your own site. What is certain is that a new layer now sits between a searcher and your page, and it decides which pages get surfaced inside it.

Does Google reward or punish AI-written content?

Neither, in itself. Google's position is about quality and purpose.

In February 2023, Google wrote that its ranking systems aim to reward original, high-quality content that demonstrates experience, expertise, authoritativeness and trustworthiness, which it calls E-E-A-T, and that its focus is on the quality of content, not how it is produced.³ It also said that using automation, including AI, to generate content with the primary purpose of manipulating rankings violates its spam policies, while not all use of automation is spam.³

The March 2024 line in the sand

In March 2024 Google announced three new spam policies, including scaled content abuse: many pages generated primarily to manipulate rankings and not to help users, "no matter how it's created."² It said the policy applies whether content is produced through automation, human effort or a combination.²

That matters for anyone tempted to ask a model for two hundred blog posts on every keyword. The same instinct that makes fake testimonials a legal problem makes fake expertise a ranking problem. The risk is not that the tool is AI. The risk is that the pages exist to rank, not to help.

How does expert content get cited?

Here is my read: the pages most likely to be quoted are the ones that make quoting easy and trustworthy.

Answer the question first

Put a direct answer near the top, in plain language, in one to three sentences. Then support it. A summarizing system has to extract an answer. If yours is buried in the fourth paragraph under a story, a competitor's tidy answer is easier to use.

Be specific

"It depends" is true and useless. "It depends on three things, and here is how to check each one" is quotable. Specifics, such as named steps, defined terms and clear conditions, are what separate an expert page from a generic one.

Show first-hand experience

The first-hand part of E-E-A-T is the experience in it.³ A page that shows how you did something, what you saw and what you would change reads differently from a rewrite of the top ten results. I do not know how Google measures it. I do know that readers can tell.

Structure the page

Use clear headings, short paragraphs, lists where the content is a list, and a short FAQ at the end with real questions. This is for humans first and machines second, and it helps both. You will notice this article is built the same way.

What should you stop doing?

• Publishing in bulk. Volume without value is exactly what the scaled content policy targets.
• Rewriting competitors. If your page adds nothing the top result does not, you are one more copy in a pile.
• Hiding authorship. Anonymous pages give a reader, and a system, less reason to trust you.
• Padding. Every extra paragraph that does not answer something dilutes the page.

Use AI as the assistant, not the author

Use a model to research, outline and tighten, the same way I suggest using it as a writer for your SOPs. Then put your own experience, claims and voice on top. The goal is a page only you could have written.

Should you panic about the 25% prediction?

No, but do not ignore it.

Gartner predicted in February 2024 that traditional search engine volume would drop 25% by 2026, as AI chatbots and other virtual agents take on some of that role.⁴ It is a prediction from an analyst firm, and the date it refers to has not arrived. Treat it as a reason to diversify, not as a certainty.

The durable response

Build an audience you own. Email lists, communities and direct relationships do not depend on which paragraph Google chooses to show. Search is a channel. Your authority is the asset, and as I said in my mid-2024 read on paid costs, no single platform should decide whether your business works.

Turning what you know into content that people and search engines trust is work I am glad to help with.

Keep building.Ekai