How to Appear in AI Search Results: Your AI Search Survival Guide
With every update, launch or AI tool, the same old Linkedin posts pop up over and over. “SEO is dead.”
It’s not, that’s a lazy rhetoric that’s been around for years. Here’s a news story from 2015. Need we say more?
But what is dead, or dying, are the outdated tactics some bad SEOs are still clinging on to.
We’re talking buying ££££££s worth of irrelevant albeit high DA links, running Digital PR campaigns with absolutely no relevance to your service/product, churning out blog after blog of generic AI-generated content (we can all see it).
Whilst these strategies can give you a pretty good traffic boost at first, they’re unsurprisingly the strategies that fall flat every time there’s a Google update.
As Charlie (Minty founder) puts it:
“SEO isn’t dead. But outdated tactics are.”
“If one Google update can wipe you out, you didn’t have a strategy. You had a shortcut.”
Google’s not the only player anymore. People are getting answers from ChatGPT, Gemini, Bing Copilot, TikTok and Reddit. That’s what we mean when we say search is fracturing. Your audience is still searching, but not just in one place, and not in the same way.
The good news? The core of good SEO hasn’t changed: solve the right problems, say something original, earn trust, and build a proper brand. The opportunity today is in showing up where your audience is looking, across every touchpoint.
That’s what this report is about. A no-nonsense look at what’s happening to search, where AI fits in, and what that means for brands in 2025.
Let’s get into it.
2. AI’s Entry into the Search Conversation
- Nov 2022: ChatGPT launches (OpenAI)
- Feb 2023: Bing integrates GPT-4 into Bing Chat (Microsoft)
- March 2023: Google announces Bard (renamed Gemini)
- April 2023: Perplexity AI launches publicly
- May 2023: Google tests AI Overviews in Search Labs
- Dec 2023: AI Overviews start rolling out to users in the U.S.
- May 2024: ChatGPT hits 5th most visited website globally (Search Engine Land)
- June 2024: Full U.S. rollout of AI Overviews (Search Engine Land)
- Ongoing: Meta and Apple exploring LLM-integrated devices and assistants (Search Engine Land)
3. AI Search Use Cases
Currently, more than half of U.S. adults (52%) are utilising tools like ChatGPT to find answers online.
A growing chunk of users (27%) are turning to generative AI for at least half of their searches. That’s a big shift in a short time.
Perplexity has quickly gained traction too, especially among tech-savvy users and academic researchers. It’s clear that tools like ChatGPT, Gemini, and Bing Copilot are becoming part of everyday behaviour, particularly for people who want faster, cleaner answers without wading through a page full of ads.
That said, AI search isn’t perfect. These tools are helpful, but they’ve got real limitations:
- Hallucinations: AI can still make stuff up. Confident, convincing nonsense is a known issue in LLMs, and it’s risky if users aren’t checking sources.
- Lack of real-time data: Many models, especially older versions, rely on outdated training data – so they miss recent updates, product launches, or news.
- Zero-click answers: AI tools often surface direct answers without sending users to websites, which can reduce site traffic even if the brand is mentioned.
- Tone and nuance: LLMs struggle to interpret sarcasm, irony, or regional slang – which affects both the way they answer and the way they represent your brand.
Some industries are flying with this tech – SaaS platforms, customer support teams, and content marketers are using AI to speed up workflows and surface insights faster. But others are lagging behind. In healthcare, legal, and anything high-risk or compliance-heavy, trust is still too important to fully hand over to a model that might get it wrong.
At the same time, users are becoming more critical of traditional search. Complaints about cluttered results pages, irrelevant answers, and inaccurate AI Overviews are growing. People want easy, fast, trustworthy answers, and they’ll go wherever they can get them.
4. Friend or Foe?
The numbers tell a more balanced story than those we see with the sole intention of fearmongering.
Google is still very much in charge, owning around 75% of search share globally. Compare that to ChatGPT, which accounts for just 2.96% of search traffic today. AI tools are growing, but they’re not replacing traditional search engines at the moment.
The truth is, most people don’t want AI to do everything for them. What we’re seeing is a shift in search behaviour. Consumers are experimenting with tools like ChatGPT, but they still lean heavily on search engines for deeper research, shopping, or anything where trust matters.
Here’s where AI does make a real impact on SEOs – it makes you more efficient.
AI tools are already speeding up:
- Content research and briefing
- Keyword clustering and internal linking
- FAQ block generation
- Competitive audits
- Prompt building for structured answers
That’s a huge win, especially for busy teams. But it doesn’t replace the need for strategy, critical thinking or experience. AI can help you do SEO faster, but it doesn’t think like an SEO. You still need to understand intent, spot technical issues, and plan long-term.
The bigger challenge is that visibility doesn’t always mean traffic anymore.
As more AI-generated results show up directly in search (like Google’s AI Overviews), we’re seeing a rise in zero-click answers, people get the information they need without visiting a site. That’s already changing how we measure performance. Brand mentions, citations, and appearances in AI answers are becoming just as important as clicks.
Despite all this change, only 22% of marketers are currently tracking visibility in AI tools like ChatGPT or Gemini, even though 82% of consumers say AI search is more helpful than traditional engines. That’s a massive gap between behaviour and strategy.
Enter Generative Engine Optimisation (GEO).
Think of GEO as the next step in SEO: making sure your content and brand are structured, mentioned, and trusted enough to show up in AI-generated answers. It’s early days, but many are already adjusting their tactics to be visible not just in SERPs, but in the results people are now getting from AI.
5. How Is AI Changing the path to conversion?
The idea of a clean, linear funnel, awareness to interest to purchase, doesn’t reflect how people search anymore. That journey is broken up, scattered across platforms, and increasingly influenced by AI.
Today, someone planning a holiday might start on TikTok, jump into ChatGPT for comparisons, check a brand’s website for prices, then head to YouTube for reviews. That’s not unusual. It’s the norm.
It’s no longer “just Google it”. It’s happening across Google, ChatGPT, TikTok, Reddit, Amazon, and YouTube, sometimes all in one decision-making process. Each platform plays a different role, and AI is making that even more fluid.
Consumers now expect:
- Instant results with minimal effort
- Contextual answers that understand their intent
- Fewer ads and less friction
- Answers where they are, not just on Google
AI tools often provide clean, direct answers without endless waffle or hard sells, especially helpful for quick comparisons, ideas, or clarifying something simple.
We’re also seeing more zero-click behaviour. People are reading summaries, getting answers, or discovering products without ever clicking through to a site. That’s happening through featured snippets, AI Overviews, and chat-style search results.
This shift has also fuelled a rise in searchless discovery, the idea that people are finding answers before they even think to ask. AI surfaces relevant topics based on context or conversation, not just queries.
It’s changing how we think about visibility. Your brand might not show up as a blue link anymore. Instead, you’re in the response a user sees from ChatGPT, or the cited source in a Google Overview. That still counts but only if you’re building authority in the places these engines are pulling from.
6. Visibility & Entities
Digital PR has always been about more than just links, at least for us.
But with AI search, its role has become even more central because AI doesn’t just read your content. It understands it by identifying what are known as entities.
Entities are the named things in a piece of content: brands, people, places, dates, even events. Tools like ChatGPT, Bing Copilot, and Google’s AI Overviews are built on knowledge graphs that map how these entities relate to each other, and those graphs are influenced by the mentions and context they pick up from across the web.
That means it’s not just about what your site says. It’s about who else is talking about you, and where.
To show up in AI-generated results, your brand needs to:
- Be cited by trusted publications
- Appear in structured, context-rich content
- Use schema markup, FAQ blocks, and clear formatting
- Be part of topical conversations that build your authority
This is where Digital PR does the heavy lifting.
Let’s take a quick example, Kinglike Concierge (KLC), a luxury villa rental brand working with Minty. As part of their strategy, we ran digital PR campaigns that:
- Earned 150+ placements in high-authority outlets like Forbes, Travel + Leisure, TimeOut, and The Express
- Used structured data, clear internal linking, and FAQs designed for AI visibility
- Covered topics tied to luxury travel, affordability, and search trends
The results? KLC appeared in AI summaries on ChatGPT, Bing Copilot, and Perplexity for queries like “best luxury travel destinations” and “where to stay in Mykonos”.
This wasn’t by accident, it was by design. Each campaign was planned not just to land links, but to create brand mentions in the kinds of sources these AI models pull from. The content was built with AI-readability in mind: clean copy, smart markup, and data-driven angles that added credibility.
So yes, Digital PR still boosts SEO as it was. But it now plays an even bigger role in how AI models decide which brands to surface. It’s one of the clearest paths to visibility that we can identify at the moment.
7. Navigating AI With Clients (and Skeptics)
It’s no surprise that clients are asking questions right now. AI is everywhere, and with it comes a flood of assumptions.
Some of the most common ones we hear:
- “Isn’t it easier now with AI?”
- “Shouldn’t this cost less?”
- “Is SEO even worth it anymore?”
They’re fair questions but they miss the bigger picture.
AI can speed things up. It can help generate outlines, cluster keywords, write drafts, spot technical issues, and support ideation. But that doesn’t mean the work is done for you. What AI gives us is scale, the ability to do more with the same amount of time, not less work overall.
If anything, expectations are rising. Brands want better, faster, more consistent content, across platforms, formats, and campaigns. AI helps us get there, but only with the right people behind the tools.
Half of marketers say they feel pressure to adopt AI. But only 11% feel over-reliant on it (Source: Search Engine Land, SMX Advanced 2025 AI Search Report.) That gap shows the reality: most teams are trying to keep up, not cut corners. And those who rely too heavily on AI without strategy end up with generic, soulless output, content that ticks boxes but doesn’t land.
We’ve seen this play out firsthand. In the KLC campaign, the client’s previous agency had filled the site with bland AI-generated copy. It was fast, yes, but it didn’t reflect the brand, it didn’t match search intent, and it didn’t build trust. We had to rip it out and rebuild with human-first content that actually resonated, and that’s what delivered results across organic search and AI summaries.
The most important thing to explain to clients is this:
The only way to keep AI from taking your job is to use AI to do your job better.
That means better briefs, stronger prompts, tighter workflows, and yes, more time to spend on strategy, creativity, and technical execution.
So no, AI isn’t making us redundant. It’s giving us new tools. The work still needs expertise. It still needs a plan. And it still needs people who know what they’re doing.
This shift also means we need to evolve how we present our work.
Clients still expect to see rankings, traffic, and links. But if that’s all you show, you’re missing half the picture, especially when your campaigns are surfacing in ChatGPT answers, Bing Copilot citations, or Google AI Overviews.
Start reporting on:
- AI visibility (tools like Quno or Rankscale can track this)
- Mentions in AI summaries across different platforms
- Citation quality: where your brand is being referenced and by who
- Sentiment shifts and brand perception from Digital PR
- Competitor presence in AI tools to benchmark your impact
Don’t rely on ChatGPT to tell you where you appear! Use tools built to monitor AI outputs. You wouldn’t use Google Search Console to run a TikTok campaign, so don’t use a chatbot to assess visibility.
In pitches and monthly reports, frame AI visibility as:
- An extension of organic performance
- A credibility signal: being cited as a trustworthy source
- A competitive advantage: most brands still aren’t optimising for AI results
The goal is to make it clear that your work is shaping how the brand is seen across search environments, including ones without links.
8. AI & Brand: Don’t Lose the Plot
Don’t lose touch with reality. ChatGPT isn’t a person on the other side of the screen, genuinely interested in your feelings or why your boyfriend said that thing he said — or how to phrase a passive-aggressive text to your mother-in-law in a slightly less passive-aggressive way. It’s just giving you statistically likely language based on its training. That’s why brands still need to bring the insight, accuracy, and experience.
One of the biggest risks in this AI boom is that brands lose their voice. When you feed generic prompts into an AI tool and publish what comes out, you don’t just end up with bland content, you lose the trust of your audience. You lose identity. You lose the stuff that actually drives people to convert.
AI-generated content still needs shaping. It still needs a clear tone, structure, purpose, and a human point of view. Without that, you’ll blend into the background, fast.
Brand tone matters more than ever. The internet is drowning with auto-generated blogs, “expert” posts, and waffle-heavy answers.
That’s why prompting matters. The quality of AI output depends entirely on how you guide it , the input, the brand context, the nuance. Vague prompts get vague results. Smart prompts, rooted in your actual tone and customer insights, get content you can work with.
There’s also the trust factor. Hallucinations, AI making things up, are still common. The best way to reduce this is to feed your models with data. In other words: be a source of truth, not just a source of content.
That means building your own bank of:
- Customer insights
- Brand research
- FAQ documents
- Tone of voice guides
- Product data
- Subject Matter Expert input
These become the core of any AI-assisted workflow. Without them, you’re guessing.
It also means keeping people informed. Human oversight, clear author bios, and editorial checks are key to building trust, with both users and search engines. If your content is supposed to demonstrate expertise, it needs to show where that expertise comes from.
As AI gets smarter, GEO reminds us that brands need to show up clearly and consistently, across formats, platforms, and outputs. Structured content, strong messaging, and recognisable tone are non negotiables.
In summary
SEO has become messier, more fragmented, and frankly, more interesting.
Your job isn’t to chase every new tool or jump on every trend. It’s to understand how people are looking for information now, and make sure your brand is present, clear, and credible wherever they go, whether that’s Google, ChatGPT, TikTok, or a well-timed travel feature.
That means doing the basics well: clean site structure, good content, useful links, earned coverage. Then layering on the new bits: AI optimisation, structured data, entity recognition, prompt-aware copy.
It’s about adapting your strategy to fit the reality of how search works today, not how it worked five years ago.
If in doubt? Go back to first principles: be useful, be findable, and sound like someone worth trusting.
