Your next social media campaign could be an AI Search result

Kinda Jackson, Managing Director Digital & Social

AI search isn’t coming, it’s here and changing fast.

We’ve already talked about how large language models (LLMs) are rewriting the rules of search and the impact that is having on traditional PR. But, that shift, away from links and keywords, toward language, context, and intent, is already re-shaping how people find answers online. But there’s a further conversation to be had: what happens when social media content becomes part of the training data?

Social is search. Especially now.

Search doesn’t start on a search engine anymore. It starts on TikTok, on Reddit or on Instagram. Ask anyone under 30 how they found a restaurant, booked a trip, or made a financial decision recently, they’ll likely show you a Reel, a comment thread, or a creator’s post.

Social platforms are no longer just places to be entertained or inspired. They’re a primary discovery tool for many and AI models understand this.

The Internet that Trains the machine

LLMs learn from the web. They process billions of datapoints, so that when people talk about your brand, your category, or your values online, those signals shape how AI understands you.

The problem is that not all signals are created equally. Models prioritise relevance, consistency and credibility, meaning the niche Reddit thread with 42 upvotes, might shape an answer more than a news article with a corporate quote.

In fact, Meta recently announced that public Instagram posts from Business or Creator accounts will now be indexed, meaning they will automatically become searchable through Google and other search engines.

With that in mind, while we know great PR will continue to build reputation, reinforce messages, and place brands in the conversations that matter, social posts do this, beyond their original platforms. Now, your social content needs to also be an SEO play, with your videos, images and posts potentially an answer to a user’s “Google search”. It puts a focus on the things we’ve always thought important: repeated mentions in the right places, consistent association with key themes and values; subject matter credibility and sharp, clean writing. It does mean that SEO, social and traditional PR need to work in combination in a way they never have. Integration between experts in each of their fields is going to be even more important, marking a shift in how you need to work to create social content too.

Communities move the needle

In tightly connected communities, whether it’s Substack, Discord, or even a very specific Instagram comment section, people share, debate and validate ideas with a level of intensity that algorithms take notice of. For LLMs, those patterns look like consensus, which gives your brand an opportunity.

Being present, useful and engaged in the right places matters more than being visible everywhere. Especially if you’re trying to shape future-facing topics like AI, sustainability, or wellbeing, where the quality of conversation matters more than volume.

This shift also re-frames how we think about influence. Reach is no longer the endgame, when reputation can be so easily impacted.

We’re seeing LLMs favour voices of domain experts, educators, and long-time creators more than those chasing virality. As a result, brands should consider social partnerships through a new lens, one that considers if the creator can build a meaningful, enduring narrative with machines, not just audiences.

Smaller creators with deep community trust are often more influential in AI terms than bigger accounts with looser followings, meaning the impact you can have with smaller creators may be greater than ever before. Essentially, it’s no longer about borrowing the reach of your favourite creator and instead is about influencing how the internet understands and describes your brand.

What brands should do next

This isn’t about rewriting your entire strategy. But it does mean reframing the role of social in your broader communications strategy. Here’s a few switches to consider:

  • Collaborate with credible creators, not just the loudest.
  • Join the right conversations, the ones most relevant to your brand and audience, not just the trending ones.
  • Track what’s being said, not just how many people are saying it.
  • Think beyond short-term clicks, ensure social content builds your brand reputation for longevity, not just using clickbait to go viral.

If AI search is now powered by the collective voice of the web, brands need to be a part of it. Those that win won’t be the ones who shout the loudest, they’ll be the ones who understand how influence travels. They need to use the people, platforms and now the models that remember everything to their advantage.

If that sounds daunting, it shouldn’t. It just means we need to think more deeply about what we say, who we say it through, and where that message might travel next. When building your next campaign, instead of focusing on how it can go viral, instead consider if it can become an answer to a search query.