Why 39% of AI Systems Disagree on Brand Recommendations: Insights from 2,881 Prompts

Why 39% of AI Systems Disagree on Brand Recommendations: Insights from 2,881 Prompts

Here’s the awkward truth most businesses haven’t wrapped their heads around yet: AI doesn’t offer universal answers. What ChatGPT recommends isn’t guaranteed to match Google’s AI results. And don’t even get me started on Perplexity or Claude. Across our dataset of 2,881 prompts, we found that 39% of the time, AI systems give conflicting brand recommendations for identical queries. Let that sink in. Nearly four out of ten searches generate divergent results depending on the AI you’re asking.

So, why’s this happening? And more importantly, what are the implications for businesses trying to be visible in this fragmented AI world?

It’s a question I’ve been obsessed with since day one at Contxt, and frankly, the deeper we dig into the data, the more complicated it gets. But that’s a good thing, because understanding these discrepancies is how brands can actually strategise for better AI visibility. Let’s break it down.

Why AI systems disagree

First things first: AI brand recommendations aren’t governed by some universal algorithm. Each AI system, whether it’s ChatGPT, Google AI, or Claude, has a unique architecture, training approach, and way of processing prompts. They’re not speaking the same language, even when answering the same question.

Here’s what I keep seeing across our experiments:

1. Training data variation: Some models are trained on vast datasets (ChatGPT, looking at you), while others, like Google AI Overview, lean more heavily on real-time web crawling. It massively impacts what they “know” about your brand.
2. Query interpretation: AI systems don’t interpret prompts identically. A query like “What’s the best industrial pump manufacturer in the UK?” may trigger brand visibility on ChatGPT but fail to on Google AI if the phrasing doesn’t align with their indexed content.
3. Biases and filters: Each AI provider applies its own filters, some explicit (like anti-spam mechanisms) and some buried in the architecture itself. These influence which brands are mentioned, especially for commercial queries.

Let’s look at the numbers to ground this. Across the prompts we tested, ChatGPT had a brand mention rate of 75.4%, the highest of any provider. Meanwhile, Google AI lagged way behind with 45.5%. Claude wasn’t far off ChatGPT at 78.5%, but Perplexity hovered at 50.0%. Related reading: The Hidden Cost of Ethical Blind Spots in LLM Visibility .... Related reading: this piece about seo is not geo: why ranking on google and getting cited b.... This isn’t just a gap, it’s a chasm.

And it gets messier. Viking Pumps, for example, hit a perfect 100% visibility rate across our network. Meanwhile, Iconsys 2, a manufacturing and industrial automation business, barely scraped 6.7%. Related reading: this piece about the llm consolidated testing framework: 36 signals, 37,00.... Related reading: this piece about the business case for llm visibility: why it’s non-negoti.... That’s not a small disparity; it’s visibility swinging between “everywhere” and “completely invisible” depending on the AI system.

The hidden cost of divergence

If you’re a business, this divergence should scare you a bit. Imagine running paid campaigns, investing in content, optimising your website, and still being invisible on nearly half of AI platforms. And it’s not just tech brands that face this problem. Entertainment, sports technology, B2B SaaS, every sector’s feeling the pinch.

Take TOCA Social, for example. Their visibility rate sits at 10%, near the bottom of our dataset. Now, TOCA isn’t a tiny player; they’re making waves in sports tech. Related reading: our article on how b2b enterprises can dominate llm visibility in ai sys.... You can visibility tracking tools to see for yourself. But if Google AI Overview users can’t find them, does it matter? A mate of mine who runs an agency put it bluntly: “If you’re invisible, you’re irrelevant.” It’s harsh, but in the context of AI, it’s spot on.

Even worse, the divergence impacts buying intent queries. These are the prompts where users are essentially saying, “I’m ready to spend my money, show me the options.” Across 12 businesses we tested, brand mention rates varied dramatically by query type. You can how AI visibility monitoring works to see for yourself. It’s not enough to rank well for generic questions; you need consistent visibility where it truly matters, at the point of purchase.

The implications for strategy

So, what’s driving this visibility gap, and how can businesses fix it? My theory? AI visibility isn’t just about SEO anymore. It’s about understanding how your brand’s footprint appears (or doesn’t) across multiple AI systems. And that requires a shift in thinking.

You can’t assume that being visible on ChatGPT means you’ll show up on Google AI Overview. In fact, our data proves the opposite, brands often perform well on one platform but poorly on another. If you’re not tracking these discrepancies, you’re flying blind.

Here are four strategies businesses can use to tackle cross-provider divergence:

1. Audit your AI presence: Start by testing how your brand shows up in AI prompts across providers. We’ve run 85 tracker experiments, and trust me, the insights are eye-opening.
2. Diversify your content formats: AI systems favour different types of content. ChatGPT might lean on in-depth articles, while Google AI prioritises live links or FAQs. Test, tweak, repeat.
3. Optimise for prompt sensitivity: This one surprised me. The same brand can be recommended for one query but absent in another, even if they’re closely related. Experiment with phrasing to uncover gaps.
4. Track updates in AI algorithms: Providers like ChatGPT roll out updates that change how they process prompts. If you’re not monitoring these shifts, you’re liable to fall behind.

Why this matters more than most people realise

At the end of the day, this divergence isn’t just academic, it’s deeply commercial. If 39% of AI systems disagree on brand recommendations, then 39% of your audience could be missing you entirely. And that’s assuming you’re doing well, several brands in our dataset didn’t even break 20% visibility.

One client I worked with knew they were underperforming but didn’t realise how widespread the problem was until we ran the numbers. They were almost entirely absent on Google AI Overview, despite holding a strong position on ChatGPT. Within six weeks, we pinpointed the issue: outdated content structures that weren’t surfacing on Google’s real-time crawlers. Fixing that helped them climb to 64% visibility across the network, better, but still room to grow.

Businesses need to be asking the hard questions. Not just “How are we ranked on Google?” but “Does AI actually know who we are?” Because visibility gaps are where competitors sneak in. If you’re absent, someone else is filling that space.

Looking ahead: What’s next?

Here’s the exciting part. AI visibility is still in its infancy as a discipline. The fact that we’re seeing data like a 39% divergence rate in brand recommendations means there’s massive room for optimisation. But it’s also a bit of a minefield. Without clear metrics and tools, brands are left guessing.

That’s why I built Contxt, to solve this exact problem. Our platform tracks how brands perform across AI systems, analyses divergences, and surfaces actionable insights. It’s not rocket science (I’m no genius), but it’s designed to close the visibility gaps that leave businesses invisible.

As AI systems evolve, staying visible across all platforms isn’t just a challenge, it’s an opportunity to get ahead of the curve. Imagine knowing that your brand ranks consistently on every major AI provider. That’s not just survival; it’s dominance.

Your move

Look, the future isn’t slowing down. AI isn’t going away, and neither are these divergences. The brands that win in the next five years will be the ones that treat AI visibility as seriously as they did SEO a decade ago. My advice? Start now. Audit your brand. Test prompts. Track visibility. Adapt.

Because here’s the thing: every day you wait, another prompt is run where your name doesn’t show up, and someone else’s does.

It’s your choice whether that’s the status quo or the wake-up call.

If you’re ready to track your AI presence and optimise for multi-platform visibility, let’s talk. The future of brand recommendations might be messy, but it’s also bright for those who act.