Why 39% of AI Systems Disagree on Brand Recommendations (And What It Means for GEO)

Why 39% of AI Systems Disagree on Brand Recommendations (And What It Means for GEO) Imagine being everywhere and nowhere at the same time. Sounds impossible, right? But that’s essentially what’s happening to brands in the world of AI recommendations. In our testing at Contxt , we discovered something that blew my mind: 39% of the time, ChatGPT and Google AI give conflicting brand recommendations. That’s not just a minor glitch; that’s a fundamental problem for businesses investing in Generative Engine Optimisation (GEO) . My take: this matters more than most people realise. We're not talking about quirks or flawed algorithms here. We're talking about whether your brand is visible to potential customers or relegated to the shadows, depending entirely on whose AI they’re using. And the implications for marketing, GEO strategy, and even competitive positioning are huge. Abstract visual of AI divergence with brands split across platforms It’s a Cross-Provider Visibility Problem Here’s what our data tells us. Across 2481 AI-generated prompts tested, we found significant variations in how major systems like ChatGPT, Google AI Overview, Claude, and Perplexity recognise brands. Let’s break down the numbers: ChatGPT delivers a 74.8% mention rate, higher than most other systems. Google AI Overview sits far behind at 44.2%, and Perplexity lags even more at 41.4%. Claude, on the other hand, smashed it with a 93.3% mention rate, but only over 30 prompts. Sounds great if you’re leveraging ChatGPT or Claude, right? But here’s the kicker: nearly 40% of prompts returned different answers depending on the AI system. A brand visible on ChatGPT might be completely absent on Google AI Overview, or vice versa. You can’t just assume you’re covered across providers. Related reading: From 0% to Cited: What Brands Scoring 80%+ on AI Actually... . Related reading: our article on the two visibility problems every business now has (and w... . Visibility gaps are real, and they’re costly. Why Ar