AI search is no longer a conversation about the future; it has become a fast reality that is reshaping search at a staggering pace. This shift, which has been driven through AI powered search such as ChatGPT, is forcing SEO professionals to adapt their strategies or risk being left behind. The data shows this too, where studies make one thing clear: the way people find information has fundamentally changed.
AI-Driven traffic surge signals a new era for SEO
In just the first five months of 2025, AI-driven discovery has grown at an unprecedented pace. According to the latest Previsible AI Traffic Report, sessions from large language model (LLM) platforms such as ChatGPT, Perplexity, Claude, Gemini, and Copilot increased by 527% compared to the same period in 2024. This is an increase from 17,076 to 107,100 sessions.
The graph below depicts this exponential growth per LLM.
While many marketers have spent the past year speculating on whether AI search would disrupt traffic patterns, the conversation has now shifted to how much it already has, often without businesses realising it. LLM’s are becoming a legitimate search platform, and fast.
The graph below depicts this growth by industry.
The recent growth, is so significant that a study conducted by SEM Rush predicts LLM traffic will completely overtake traditional Google search by 2027.
This growth is not limited to OpenAI’s ChatGPT, although it remains the largest driver (responsible for 40–60% of AI-driven sessions). Perplexity, Copilot and Gemini are gaining a lot of traction. This suggests we are moving towards a multi-model environment where visibility across platforms will matter just as much as traditional search rankings.
AI Search as a strategic discovery channel
The shift to LLM-based search may be familiar to some, for example, think back to the rise of social media as a search platform. It was the early adopters of this change who succeeded, and it will be the early adopters of AI who will enjoy the same first-mover advantage.
The biggest difference between these two significant changes in search, is that LLMs are not restricted to Google’s crawl–index–rank cycle. They surface content instantly if it is (1) relevant, (2) well-structured, and (3) trustworthy. This means content does not need to rank in the conventional sense to be found. Instead, it needs to be selected, and that selection happens based on perceived authority, clarity, and the model’s internal trust signals.
For instances where users ask complex, contextual questions eg. “What clauses should I check before signing a lease?”, “What are the safest investment options for retirees?”, “How should a small retail business handle payroll?”, AI is becoming the go-to search engine.
Reddit’s move into AI-Powered answers
This AI evolution is not limited only to AI assistants. Reddit, a platform known for its human-driven content, is now leaning into this new era. With over 70 million people using its internal search weekly, Reddit is heavily investing in its new AI-powered “Reddit Answers” feature, which has seen a 5x increase in its weekly users. The company is positioning itself as a go-to search destination for “a breadth of conversations and knowledge you can’t find anywhere else.”
This strategic move by Reddit just highlights the significant changes happening in search right now due to AI.
What can you do to adapt to these changes?
If you are still treating AI-driven search as a future problem, you are already behind. The focus now must be on building the kind of presence, both on and off your site, that LLMs and AI-powered search experiences want to reference.
Here are actionable steps to adapt:
1. Track LLM-Driven traffic:
Even imperfect attribution is better than none. Use UTM tagging for known AI referrers, monitor unexplained direct traffic spikes, and annotate when your content is surfaced in AI responses.
Your brand might be blowing up on LLM’s without you even realising it. The fastest way to find this out, is by looking at the data.
If your GSC is showing a pattern similar to that of the screenshot below, where your clicks have dropped yet your impressions have surged over a period of time, this generally indicates that more people are seeing your content in search results, but fewer are clicking through to your site. In other words, they’re likely discovering you through AI responses and then searching for you directly later.
2. Structure your content for AI selection:
Use concise summaries, clear headings, bullet points, and FAQ sections to make your content easily “skimmable”. AI models prefer structured, scannable answers over long, unformatted content. The result is content that’s AI-ready, discoverable, and created for selection in conversational search interfaces.
3. Build authority across platforms
Secure features on respected and authoritative third-party sites, participate in authoritative forums eg. Reddit, and ensure consistent branding across the web. LLMs frequently pull from trusted external sources, so being present on those platforms increases your indirect AI citation potential.
Another strategy to build authority for your site is through creating content around E-E-A-T (Experience, Expertise, Authority, Trustworthiness). By creating content around staff members’ expertise, expanding on you “About us” pages, including testimonial pages with reviews, this signals to AI platforms that you are an expert in your space.
4. Optimise for “Multi-Model Discovery”:
Different AI platforms prioritise different sources, for example, Perplexity favours expert forums, ChatGPT leans on authoritative articles, Gemini blends news and community input, and Reddit Answers sifts through it’s human discussion archives. Understand these preferences and tailor your content strategy accordingly.
We have a great article that dissects this entire topic and approach. Click here to read more on it.
Why we care
SEO is no longer just about optimising for Google’s blue links. It is very quickly splitting into two parallel paths: traditional search and AI-powered search.
The winners in this shift will be the brands that move early, build trust signals across multiple authoritative environments, and adapt content for AI’s specific ranking criteria. With platforms like Reddit focusing on AI answers and LLM traffic already showing triple-digit growth, the path forward is clear: understand where AI is sourcing information, position your brand to be part of those sources, and optimise for the way content is cited on these conversational platforms.