As AI such as ChatGPT spreads, companies that have used search-engine optimisation (SEO) for more than 20 years to win consumer attention are looking for a range of alternatives.

That is driving growing interest in generative engine optimisation (GEO) and answer engine optimisation (AEO).

A McKinsey survey in August 2025 found half of respondents actively use AI-based search engines. McKinsey forecasts users will spend $750 billion by 2028 on products and services found through AI-based search.

AI chatbots such as ChatGPT and Claude provide answers directly to questions. For companies, it is becoming important not only to rank high in search results but also to be a source that chatbots use. But on the AI side there are not yet clear standards or know-how companies can follow like Google search. It is also not easy to figure out which sites AI uses and recommends when it answers questions.

As more experts dig into how AI works, the veil is gradually lifting on how to become information that AI finds.

A recent Wall Street Journal report said advice from experts on this can be summarised in a few points.

There is no need to break with search because AI is rising. AI is growing quickly, but many people still use traditional search engines and click links. Even when users use AI chatbots, web search takes place in the background. Kelly Cutler (켈리 커틀러), a professor of digital marketing and visual communication at Northwestern University, said, "AI is not the end of SEO. Next-generation SEO."

Even so, the rules of the game are changing in marketing. Under traditional SEO, getting links from authoritative sites was favoured in rankings over being mentioned hundreds of times on other sites. But AI chatbots are different. The WSJ said AI favours the relative volume of customer reviews and comments drawn from sites such as Reddit, G2 and Quora.

Andrew Warden (앤드류 워든), chief marketing officer at digital marketing company Semrush, said, "AI search is inseparable from user-generated content, what people say about brands." He said, "AI prefers information based on direct experience, specificity, and stories that are continuously updated. The more activity around a brand, the higher the likelihood that the brand is included in the conversation."

To increase such exposure, the WSJ said companies are likely to expand online engagement with customers by directly answering customer questions or encouraging customers to leave honest reviews.

It is also important to have a structure that AI can easily pull into answers. That means companies need to revisit their websites for AI. Traditional SEO strategies recommend using many subheadings in content. The WSJ said AI chatbots also require bullet-point lists, FAQs and a 'TL;DR' (Too Long; Didn't Read) summary that quickly organises key points.

Context is also an important factor. When users ask an AI chatbot to find something, they often ask with context. AI is therefore likely to prefer information in which contextual content is added to product listings.

In traditional SEO, what worked on Google also worked on other search engines such as Bing and Yahoo. SEO techniques that worked well in one industry could also be extended to other fields without major problems. But with AI, such rules no longer seem to apply.

James Cadwallader (제임스 캐드월라더), co-founder and CEO of AI search optimisation company Profound, said, "Each platform weights elements differently." He said, "For example, Gemini pulls far more YouTube results than ChatGPT and uses a broader range of sources than other AI chatbots."

Traditional search engines placed relatively less weight on blog posts or articles written through companies' paid sponsorships, but AI is different. Rachel Klein (레이첼 클라인), senior vice president at digital marketing agency Wpromote, said, "AI seems to treat this content on the same level as other articles."

Some analysis also suggests owned media, such as corporate websites and company blogs, is becoming more important in the AI era.

A recent Axios report citing an analysis by The Penta Group said about 60 percent of content cited by LLMs is proprietary content produced directly by companies. The share of traditional media is about 25 percent, with the rest made up of user-generated content and third-party platforms. It said earned media accounts for 28 percent and owned content for 22 percent, figures that differ significantly from traditional search.

Notably, LLMs prefer content closer to journalism than marketing copy. Clear headlines, an FAQ format, named authors and regular updates are cited as factors that increase the likelihood of being cited by LLMs. The Penta Group said content structure and freshness are variables that communications teams can directly control.