With AI systems increasingly drawing from credible media sources, earned media and media relations are beginning to matter more than ever these days. But just showing up in searches is not enough any more, you need to control how you show up when large language models (LLMs) pull together large volumes of data from various sources and create their own authoritative-sounding responses to offer the user.
For PR professionals used to tightly controlling their brand narrative, this challenge is so profound a new term has been coined for it: AI brand drift. So what exactly is it, and how can PR professionals retain control of brand sentiment?
What is AI brand drift?
Put simply, it is the loss of control over how a brand is represented. Sometimes, generative AI will surface fragmented, out-of-context or even negative information that has been scraped from numerous data sources online, and does not accurately reflect the brand’s positioning or values.
Unlike traditional search, where users are likely to see multiple sources featuring your brand, AI condenses everything into a single result. If that response leans on old news stories or misinformation, then your brand reputation is being defined by forces outside of your control. This makes brand stewardship more important than ever.
Tighten narrative control
In order to manage AI brand drift you need to make sure that your core message is crystal clear. Standardising the message is your main goal, and this is best achieved by pushing out clear and repeatable statements about your values, mission and market positioning. It’s also important to regularly review and refresh messaging on a regular basis, because generative AI systems are constantly scraping new data, and you have to make sure you have an output that reinforces the story you want to tell.
Monitor sentiment in real time
With brand sentiment so heavily influenced by these AI summaries, it is even more critical that you monitor relevant channels. This extends beyond traditional media tracking to social platforms, forums, review sites and even niche communities. AI systems regularly scrape from platforms like Reddit, Glassdoor and YouTube, so establishing a routine to audit these channels allows you to red-flag any outdated or misleading narratives before they’re spread online and scraped by AI. Understanding where the red flags are is your first step towards correcting them. Next you might need, for example, to contact a journalist to correct an error, release a statement or comment on a forum thread that concerns your brand, or update a webpage to ensure accuracy. The goal is to quickly address misinformation to safeguard the brand’s online reputation, ensuring the AI outputs reflect accurate and trusted information.
Guide employees to be brand advocates
Employees are your first line of defence but can also be a source of risk. With AI, casual social posts, jokes or comments on public forums can be more rapidly surfaced and taken out of context, so brands need to have strong internal policies around online commentary. This kind of effective guidance will help employees understand their role as brand guardians, and to realise that anything placed in the public domain can influence the way that AI depicts the company. By encouraging employees and making it as easy as possible for them to share positive comments about the brand, you can create a buffer against negative or distorted narratives. You should maintain a long-term view of your brand’s digital footprint, as every piece of content has the potential to leave a lasting effect.
Align marketing and PR
AI brand drift does not respect departmental boundaries, so close collaboration with marketing teams is essential. Sharing messaging bibles and content calendars can help control the tone and narrative, ensuring each touchpoint reinforces the same brand story. This can help prevent inconsistencies that AI could magnify and ensures a cohesive voice.
Engage in Generative Engine Optimisation (GEO)
GEO involves actively shaping the sources that AI is likely to scrape. Publishing high-quality, authoritative content is essential for brand visibility, but it should also be structured so that your brand’s tone and narrative is explicit. AI models pick up on repeated themes, stylistic patterns and phrasing, and it’s another reason why a consistent message is so essential in minimising the risk of brand drift. Your brand voice should be clearly embedded in each piece of content.
Keeping control of the narrative
PR and communications have always been about shaping perception, but with AI brand drift, the stakes are simply higher. Even though AI responses are generated individually, the same underlying AI model can be used across multiple apps, chatbots and websites, meaning that a misleading narrative can quickly proliferate across platforms and reach a far wider audience. AI, therefore, increases the need for brand stewardship, and brand drift is now the reputational risk that every company must prepare for.
Claire Snook is digital comms manager at strategic communications agency Ambitious.