Investment Management AI Visibility Report (Q2 2026)

Highlights

• The dataset: Over 240,000 AI source citations underpin this analysis, produced by prompts specific to the investment management industry. The AI responses from four AI models span 12 countries over the course of Q2 2026.
• Reddit surged to #1 in the back half (+7x share): The community giant overtook the directory sites that led the first half.
• 21% of ChatGPT's citations are from user-generated domains: Reddit, Wikipedia, and social, by far the highest UGC share of any model.
• 55% of Google AI Overview's citations are from commercial domains: The most brand-direct model, it will quote the firm's own page.
• ~10% average overlap between any two models: Any two of the four share only about a tenth of their top-20 cited investment-management domains, one of the lowest overlaps we measure.
• 46% of Microsoft Copilot's citations are from editorial domains: Comparison and review portals, nearly half its citations, the most of any model.

Which sources should you target to get cited as an investment management brand?

Each model's distinctive top sources, and the citation character they add up to:

How have AI source rankings changed over time in the investment management industry?

We split the observation window at its midpoint (early May 2026) and compared each domain's citation share in each half. The ranking is far from static.

Reddit surged to the #1 most-cited domain in the second half, growing its citation share more than sevenfold, while the directory and ranking sites that dominated the first half slid: Citywire fell about 38 places in share rank, Dakota about 41, and Chambers about 29. Track the current picture in the live AI visibility rankings.

What type of content do AI models cite for investment management firms?

The models differ not just on which domains but on what type of content they trust. We classified every cited domain into five categories.

Google AI Overview is the most brand-direct, sending 54.7% of its citations to commercial domains, it will quote the firm's own page. Microsoft Copilot is the most editorial, with 46.4% going to comparison and review media. ChatGPT is the crowd-and-reference model: 21.0% of its citations go to user-generated content like Reddit and social, and another 10.4% to reference content like Wikipedia, far more than any other model on both counts. A residual of 6.9%–9.8% per model hit uncategorized domains.

Do AI models cite local-language content for investment management firms?

Every English-language prompt (US, UK, Canada, Australia) returned sources entirely on English or generic domains, local-country ccTLDs accounted for 0% of citations there. Across all non-English prompts, only 50.4% of citations went to a local ccTLD, and the rate varies sharply by language.

Dutch, Swedish, and German investment-management questions stay strongly local; Spanish questions are pulled toward .com finance portals and away from any single country's domain, unsurprising given the Spanish prompts span Spain, Mexico, and Argentina, fragmenting the "local" signal across three ccTLDs. This is a ccTLD proxy, not true language detection, so these figures should be read as a directional floor.

Which AI model relies most on local sources for investment management firms?

The localization gap between models is real and consistent. Microsoft Copilot is the heaviest user of local sources in most markets, while Grok most often defaults to global English-language and .com domains, most visibly in Spanish, where only 24.9% of Grok's citations carry a local ccTLD.

In Dutch and German, Microsoft Copilot goes almost entirely local, 95.6% local in Dutch and 93.3% in German, while Google AI Overview reaches 87.3% in Dutch. In Spanish, every model cites local sources sparingly, with Grok and Google AI Overview below 31%. Google AI Overview returned no cited sources for French-language investment-management prompts in this window, so that cell is blank rather than zero.

Should investment management firms optimize for each AI model separately?

Yes, almost entirely. We built each model's most-cited domains for investment management and compared every pair to see how many sources they share. The average overlap is 10.2%, meaning 89.8% of top-cited sources differ between any two models, one of the lowest overlaps we have measured in any industry.

Google AI Overview and Grok are the closest cousins, sharing 6 of their top domains (17.6%), both leaning on youtube.com, global.morningstar.com, and unbiased.co.uk. At the other extreme, ChatGPT and Microsoft Copilot share just 5.3%, two domains out of forty, with ChatGPT leaning on reddit.com and en.wikipedia.org while Microsoft Copilot prefers regional comparison portals like notiziefinanza.it and iex.nl.

A firm visible on one model can be absent from roughly nine-tenths of the sources another model cites.

How many sources does each AI model cite per answer?

The models differ by nearly 8x in how many sources they pull into a single answer.

Grok cites 54.0 sources per response, about 7.8x Microsoft Copilot's 6.9, and a single Grok answer reached as many as 125 cited sources. Google AI Overview and ChatGPT sit close together in the middle (both about 11), though ChatGPT is far more variable. Grok offers many more slots per answer but each carries less weight, while Microsoft Copilot's short citation lists make every included source disproportionately valuable.

Context

This analysis draws from Temso's AI visibility monitoring platform, which tracks how brands appear in AI model responses across ChatGPT, Microsoft Copilot, Grok, and Google AI Overview. The dataset covers investment-management prompts, choosing, comparing, and recommending wealth managers, asset managers, and independent financial advisors, across 12 countries and 7 languages over roughly sixteen weeks (March–June 2026).

It measures what AI models cite, not whether those sources are accurate or whether the recommendations are good. A citation is not an endorsement. These findings reflect citation behavior in the monitored markets, weighted toward English-speaking and Western European/Latin American countries, and may not generalize to all firms, prompt types, or other verticals.

Methodology

How we measured this

We tracked four AI models responding to investment-management prompts in local languages across 12 countries and 7 languages over roughly 15.8 weeks. Each response was parsed to extract source citations: the URLs, domains, and metadata referenced. Domain categories (commercial, editorial, UGC, reference, institutional) were assigned from a global domain registry. Cross-model overlap was measured using each model pair's top-20 most-cited domains. Temporal analysis split the observation period at its midpoint and compared each domain's citation share in each half.

The analysis covers over 240,000 cited sources from 22,381 AI responses. Sample sizes exceeded recommended thresholds for every reported finding. Localization is measured by a country-code-domain proxy (the source records carry no detected content language), so those figures are directional and exclude English-language prompts, which sit overwhelmingly on generic domains. Model coverage varied across the observation window, so temporal movements are reported as relative ranks within each half.

Frequently asked questions

Do AI models cite different sources for investment-management recommendations?

Yes, dramatically so, more than in almost any industry we measure. Any two of the four models share only about 10% of their top-20 cited domains, so roughly nine in ten sources differ. ChatGPT and Microsoft Copilot overlap the least, at 5.3%.

Which AI model cites the most sources per answer for investment management?

Grok, by a wide margin, about 54 cited sources per answer, versus roughly 11 for Google AI Overview and ChatGPT and 7 for Microsoft Copilot. That 7.8x gap means each source in a Microsoft Copilot answer carries far more weight.

What kinds of websites do AI models cite for investment management?

It varies by model. Google AI Overview is the most commercial (54.7% of its citations), Microsoft Copilot the most editorial (46.4% comparison and review media), and ChatGPT the most community-and-reference (21.0% user-generated content plus 10.4% reference sites like Wikipedia).

Does writing in the local language help wealth managers rank in AI?

For some models and markets. Across non-English prompts, only about 50% of citations went to a local-country domain. Dutch, Swedish, and German questions stay strongly local (68–83%), while Spanish skews toward global .com portals (~31%). English-language markets pull essentially no local-ccTLD sources.

Which AI model relies most on local sources for investment management?

Microsoft Copilot. It cites local-country domains hardest across most European markets, 95.6% in Dutch and 93.3% in German, while Grok stays on global English sources, bottoming out at 24.9% local in Spanish.

How fast do AI investment-management source rankings change?

Quickly. Over roughly sixteen weeks, Reddit surged to the most-cited domain while directory sites like Citywire and Dakota fell 30–40 places in share rank. AI source rankings churn fast.

Why do directory and ranking sites like Citywire and Chambers matter so much?

They are central to Grok, which floods its answers with professional rankings and league tables, Citywire, Morningstar, Dakota, Chambers, and RankiaPro all sit near the top of its list. Earning strong profiles on those directories is the most direct lever for Grok visibility.

Is one AI ranking enough to optimize for in investment management?

No. The models share only about a tenth of their top sources and weight them very differently, so visibility on one model does not transfer to the others. An AI-visibility strategy for wealth and asset managers has to be built model by model.