UK AI Employer Visibility Index 2026
We audited 580+ UK employers. Here’s what AI gets wrong about them.
580+
Employers audited
40/100
Average score
£18K
Average salary error
58%
Hallucination rate
Key findings
78% of AI salary estimates were inaccurate — most underestimated by £15–25K, costing employers qualified candidates who assumed the pay was uncompetitive.
91% of UK employers have no llms.txt file, leaving AI models to assemble employer narratives from Reddit threads and Wikipedia stubs.
67% of companies lack structured data (JSON-LD) on their homepage, forcing AI to guess basic organisational facts.
43% of employers actively block AI crawlers via robots.txt, making themselves invisible to the platforms candidates use most.
Only 4% of UK employers score above 80/100 for AI visibility — the gap between leaders and laggards is enormous.
Companies with published salary data see AI salary deviations of just £3,200 vs £18,400 for those without.
Industry breakdown
Average AI Visibility Score by sector (out of 100)
Data from OpenRole audits of 580+ UK employers, February 2026.
Methodology
Each company was audited across four AI models — ChatGPT, Perplexity, Claude, and Gemini — using standardised employer-related prompts covering salary, culture, benefits, and interview process.
Responses were scored on accuracy, completeness, sentiment, and consistency. Technical signals (structured data, llms.txt, robots.txt, careers page accessibility) were verified programmatically. The methodology is fully documented on our scoring methodology page.
Download the full report
Enter your work email to get the complete dataset, all 580+ employer scores, and actionable recommendations.
Free, no card required. We only use your email for this report.
Where does your company rank?
Run a free audit in 30 seconds. See what AI says about you — and how to fix it.
Run your free audit