What we measured, and why banks?
We measured how often five AI engines name specific UK banks in response to 500 real consumer prompts, scoring presence, position and sentiment. Banking is universal, highly branded and increasingly researched through AI, which makes it a clear test of whether AI assistants concentrate consumer attention on a few providers.
The backdrop to this study is a broader shift towards AI-first research: over half of UK adults (54%) now use AI tools, and 47% would use AI to research a purchase (Ofcom, 2025). When someone asks an assistant “what’s the best current account for switching”, the answer they get is shaped by which brands the model has learned to associate with trust, value and relevant features, not by who has the biggest marketing budget that quarter.
Which UK banks do AI assistants recommend most?
The table below ranks the UK banks most often named across all five engines by their AI Visibility Score, combining presence, position and sentiment. Higher means named more often, higher up, and more favourably.
| Rank | Bank / provider | Presence (% of answers) | Avg. position | Sentiment | AI Visibility Score |
|---|---|---|---|---|---|
| 1 | <!-- DATA: bank --> |
<!-- DATA: % --> |
<!-- DATA --> |
<!-- DATA --> |
<!-- DATA --> |
| 2 | <!-- DATA: bank --> |
<!-- DATA: % --> |
<!-- DATA --> |
<!-- DATA --> |
<!-- DATA --> |
| 3 | <!-- DATA: bank --> |
<!-- DATA: % --> |
<!-- DATA --> |
<!-- DATA --> |
<!-- DATA --> |
| 4 | <!-- DATA: bank --> |
<!-- DATA: % --> |
<!-- DATA --> |
<!-- DATA --> |
<!-- DATA --> |
| 5 | <!-- DATA: bank --> |
<!-- DATA: % --> |
<!-- DATA --> |
<!-- DATA --> |
<!-- DATA --> |
| … | <!-- DATA: extend to top 10–15 --> |
Caption: <!-- CONFIRM: exact sample size + date range -->. Scores measure AI visibility, not product quality; this is not financial advice.
Results by question type: current accounts, savings, switching, business
AI recommendations vary by what the consumer asks, so we broke results down by prompt category.
| Question category | Most-named provider | Presence | Notable pattern |
|---|---|---|---|
| Best current account | <!-- DATA --> |
<!-- DATA: % --> |
<!-- DATA --> |
| Best savings account | <!-- DATA --> |
<!-- DATA: % --> |
<!-- DATA --> |
| Switching / best to switch to | <!-- DATA --> |
<!-- DATA: % --> |
<!-- DATA --> |
| Business banking | <!-- DATA --> |
<!-- DATA: % --> |
<!-- DATA --> |
<!-- CONFIRM: category splits after study runs -->
Methodology
We ran real consumer prompts across five AI engines from a UK location and logged every named bank, its position and its sentiment.
- Engines (5): ChatGPT, Perplexity, Claude, Google Gemini, Google AI Overviews.
- Prompts: real consumer questions across current accounts, savings, switching and business banking (for example “best UK current account for [need]”, “which bank should I switch to”, “best business bank account UK”).
<!-- CONFIRM: publish full prompt list --> - Location: United Kingdom.
<!-- CONFIRM: locale/settings --> - Dates: . Prompts distributed to reach 500 total and reduce variance.
- Scoring: presence, position and sentiment combined into a 0–100 AI Visibility Score.
<!-- CONFIRM: weighting --> - Tooling: AI-answer tracking via NeuralGen’s SE Ranking account plus manual verification.
<!-- CONFIRM -->
Why do AI engines name some banks far more than others?
AI engines lean on strong, consistently-described entities and the sources they trust: comparison sites, financial press, review platforms and reference-grade sources. Brand size helps but is not decisive. A challenger bank with heavy positive coverage can outrank a larger incumbent in AI answers.
This is a genuinely different competitive dynamic to traditional financial-services marketing, where media spend and branch footprint have historically dominated. In AI answers, what seems to matter more is how clearly and consistently a bank’s products, rates and eligibility criteria are described across the sources an engine already trusts. Read more on entity SEO and where AI answers come from.
Important limits: this is not financial advice
This study measures AI visibility, not the suitability, safety or value of any account. AI engines can repeat outdated rates, miss newer providers, and vary by phrasing and date.
Nothing in this article is a recommendation to choose any bank. Always compare current terms directly with providers or an independent comparison service, and check that any account is protected under the Financial Services Compensation Scheme (FSCS) before opening it. We keep this caveat prominent because banking is a regulated topic and the stakes of a wrong assumption are real.
What this means for banks and fintechs
If consumers ask AI which bank to use and your brand is absent or poorly described, AI is steering customers elsewhere. The fix is entity and citation work so AI engines find, trust and describe you accurately.
That includes: making sure your product details, rates and eligibility criteria are described consistently across your own site and the third-party sources AI engines cite; earning coverage on the comparison sites and financial press these engines already trust; and structuring your content so a model can extract accurate facts about you rather than guessing.
See our AI SEO agency page for how this work is scoped for regulated brands.
Will this study be repeated?
AI answers shift as models retrain and re-crawl, so we intend to re-run this study and republish the table with a fresh date. Treat any single run as directional.
We expect to repeat this at <!-- CONFIRM: annual/quarterly? --> intervals. Check the “last updated” date on this page before citing a specific figure.