Mortgage Rates: 2026 AI Market Discovery Index

Answer Capsule

In the Mortgage Rates category for May 2026, AI systems are concentrating borrower recommendations around a single dominant lender. Rocket Mortgage leads with 26.2% recommendation coverage and a 19.9% rank-one rate. Better Mortgage emerges as the strongest challenger with 9.5% recommendation coverage and the highest net sentiment in the category at 0.82. Several lenders including Own Up and Rate appear in AI responses but receive zero or near-zero recommendation credit, exposing a significant gap between visibility and shortlist eligibility.

Executive Summary

AI systems evaluating mortgage lenders are producing a highly concentrated recommendation landscape. Rocket Mortgage appears in 42% of all AI observations and earns valid recommendation credit in 29.5% of them, capturing a modeled monthly recommendation value of $261,618 out of the $369,830 total opportunity across the three public clusters. No other lender comes close to this level of AI shortlist dominance.

Better Mortgage presents the most credible challenger signal. Despite appearing in only 11.9% of observations, it earns recommendation credit in 9.5% of them and achieves the highest net sentiment score in the category at 0.82. New American Funding and loanDepot also show meaningful recommendation coverage above 6%, but both trail significantly in rank-one positioning and captured value.

The most striking pattern is the gap between visibility and recommendation power. NerdWallet appears in 25.7% of all observations, more than any lender except Rocket Mortgage, yet earns recommendation credit in only 2.3% of them. LendingTree appears in 12.6% of observations but earns credit in just 1.5%. Own Up appears in 3.4% of observations and receives zero recommendations across all platforms and clusters. These brands are being seen but not advanced, which means they are losing the shortlist battle where AI systems act as de facto buyer advisors.

The commercial stakes are real. Lenders that earn consistent recommendation credit are capturing modeled monthly AI opportunity value that compounds as borrower reliance on AI platforms grows. Lenders that remain visible but unrecommended are building awareness for competitors rather than themselves.

The AI Discovery Shift in Mortgage Rates

Mortgage borrowers increasingly use AI platforms as their first stop for lender discovery, rate comparisons, and provider evaluation. When a borrower asks an AI system for the best mortgage lender or a rate comparison, the response functions as a pre-filtered shortlist. Being mentioned is no longer enough. The question is whether the AI system recommends the lender, ranks it highly, and frames it positively.

The data shows that AI platforms are not treating all lenders equally. Rocket Mortgage is consistently ranked first across multiple platforms and prompt types. Better Mortgage earns high sentiment but appears less frequently. Several well-known brands are listed neutrally or omitted from recommendations entirely. This creates a two-tier market where a small number of lenders capture the AI recommendation opportunity and the rest are left with visibility that does not translate into buyer consideration.

Recommendation credit is earned through the quality and structure of the evidence that AI systems can retrieve. Official brand content, comparison articles, review coverage, community discussions, and regulatory or trust signals all contribute to the evidence layer that determines whether a lender is advanced or merely listed. Lenders with dense, authoritative public evidence earn shortlist placement. Lenders without it appear as neutral references.

The commercial implication is direct. AI discovery is becoming a primary channel for mortgage shopping. Lenders that do not earn recommendation credit are effectively invisible at the moment of decision, regardless of brand recognition or marketing spend.

Directional Category Leaders

1. Rocket Mortgage

Rocket Mortgage leads the category by a wide margin. It appears in 42% of all observations and earns valid recommendation credit in 29.5% of them. Its rank-one rate of 19.9% is nearly eight times higher than the next closest lender, and its average recommended rank of 1.32 means it is almost always listed first when recommended. Across the six AI platforms tested, Rocket Mortgage holds the highest recommendation coverage on Gemini (52.2%), Copilot (45.7%), and Google AI Overviews (23.3%). Its modeled monthly captured recommendation value of $261,618 represents 70.7% of the total public opportunity.

The public interpretation: Rocket Mortgage has achieved dominant AI shortlist status across all major platforms and buyer stages, capturing the majority of the category's AI-driven borrower opportunity.

2. Better Mortgage

Better Mortgage earns recommendation credit in 9.5% of observations with a net sentiment score of 0.82, the highest in the category. Its average recommended rank of 1.76 is strong, and it achieves a 6.2% top-three rate. Better Mortgage performs particularly well on Perplexity, where it captures $20,873 in modeled monthly value, and on Gemini, where it captures $22,022. Its recommendation coverage is concentrated in the discovery cluster, where it earns 10.7% of observations.

The public interpretation: Better Mortgage is the strongest challenger with high recommendation quality, but its lower overall presence limits its total captured value relative to Rocket Mortgage.

3. New American Funding

New American Funding appears in 17% of observations and earns recommendation credit in 10% of them, giving it the third-highest recommendation coverage in the category. Its net sentiment of 0.64 is solid, and it achieves a 4.6% top-three rate. The lender performs best on Google AI Mode, where it earns 14.9% recommendation coverage, and on Google AI Overviews, where it earns 7.5%. Its average recommended rank of 1.85 indicates it is typically listed second or third when recommended.

The public interpretation: New American Funding has established consistent AI recommendation presence, particularly on Google platforms, but lacks the rank-one dominance needed to challenge Rocket Mortgage directly.

4. loanDepot

loanDepot earns recommendation credit in 6.3% of observations with a net sentiment of 0.66. It achieves a 4.6% top-three rate and a 2% rank-one rate. Its strongest platform is Copilot, where it earns 27.6% recommendation coverage and a rank-one rate of 8.6%. loanDepot captures $23,906 in modeled monthly value, placing it fourth in the category.

The public interpretation: loanDepot shows meaningful AI recommendation strength on Copilot but lacks the cross-platform consistency of the top three lenders.

5. NerdWallet

NerdWallet appears in 25.7% of observations, the second-highest presence in the category, but earns recommendation credit in only 2.3% of them. Its net sentiment of 0.18 is low, and its rank-one rate is just 0.9%. NerdWallet performs best on Perplexity, where it earns 5.4% recommendation coverage, and on Google AI Overviews, where it earns 1.3%. The vast majority of its appearances are neutral references rather than positive recommendations.

The public interpretation: NerdWallet has high AI visibility as a reference source but is rarely recommended as a lender, creating a significant gap between awareness and shortlist inclusion.

6. LendingTree

LendingTree appears in 12.6% of observations but earns recommendation credit in only 1.5% of them. Its net sentiment of 0.21 is low, and its rank-one rate is 0.7%. LendingTree captures $6,443 in modeled monthly value, with its strongest performance on Gemini, where it earns 8.7% recommendation coverage. Like NerdWallet, LendingTree is more likely to be mentioned neutrally than recommended positively.

The public interpretation: LendingTree is visible in AI responses but is not earning the recommendation credit needed to influence borrower shortlists.

7. Rate

Rate appears in 8.5% of observations but earns recommendation credit in only 1.2% of them. Its net sentiment of 0.17 is low, and it achieves a rank-one rate of just 0.1%. Rate captures only $711 in modeled monthly value. The lender is mentioned neutrally in most cases and rarely appears as a recommended option.

The public interpretation: Rate has limited AI presence and near-zero recommendation power, making it a marginal player in AI-driven borrower discovery.

8. AmeriSave Mortgage

AmeriSave Mortgage appears in 3.8% of observations and earns recommendation credit in 1.6% of them. Its net sentiment of 0.41 is moderate, but its total captured value of $630 is among the lowest in the category. AmeriSave performs best on Gemini, where it earns 2.2% recommendation coverage, but is absent from Google AI Mode and Google AI Overviews entirely.

The public interpretation: AmeriSave Mortgage has minimal AI recommendation presence and is not competing effectively for AI-driven borrower attention at scale.

9. Own Up

Own Up appears in 3.4% of observations and earns zero recommendation credit across all platforms and clusters. Its net sentiment is 0.0, and it has no rank-one appearances. Own Up is mentioned neutrally in a small number of responses but is never recommended as a lender option.

The public interpretation: Own Up is visible in AI responses but receives no recommendation credit, meaning it does not appear on any AI-generated borrower shortlist.

The Buying Moments That Now Decide the Category

Discovery and Ranking

This cluster represents borrowers searching for the best mortgage lender or top-rated options. It accounts for 422 observations and the largest share of modeled opportunity value. Rocket Mortgage dominates with a 46.7% top-three rate and a 36% rank-one rate, capturing $203,814 in modeled monthly value. Better Mortgage earns 10.7% recommendation coverage here, and New American Funding earns 6.9%. This is the cluster where shortlists are formed. The data shows Rocket Mortgage is the default first recommendation across most platforms at this stage.

Comparison and Evaluation

This cluster represents borrowers comparing lenders head-to-head and accounts for 225 observations. NerdWallet leads with 5.8% recommendation coverage but captures only $1,840 in modeled monthly value. No other lender earns meaningful recommendation credit here. The low captured value suggests AI systems are more likely to provide neutral lists in comparison prompts than to make ranked recommendations, which reduces the commercial value of this cluster relative to discovery.

Pricing and Cost Evaluation

This cluster represents borrowers evaluating mortgage rates and plan costs across 240 observations. Rocket Mortgage leads with 15.8% recommendation coverage and a 10.4% rank-one rate, capturing $57,804 in modeled monthly value. Better Mortgage earns 5.4% recommendation coverage, and NerdWallet earns 2.5%. Rocket Mortgage maintains its lead even in pricing-focused prompts, where borrowers are comparing costs rather than brand reputation, which signals broad recommendation authority rather than niche strength.

Why Recommendation Power Is Concentrating

AI recommendation power in mortgage rates is driven by the evidence layer that AI systems use to construct responses. Rocket Mortgage benefits from a dense citation architecture that includes official brand content, comparison articles, review coverage, and community discussions. This public source evidence gives AI systems multiple signals to retrieve, compare, and trust the lender as a recommended option rather than a neutral reference.

Better Mortgage and New American Funding have built similar but narrower evidence layers. They appear in enough high-quality sources to earn recommendation credit, but their citation density is lower, which limits their rank positioning and cross-platform consistency. Their recommendation coverage reflects the strength of a specific source layer rather than broad platform authority.

Lenders like Own Up and Rate appear in AI responses primarily through neutral references in comparison articles or rate tables. These references are sufficient for visibility but not for recommendation credit. AI systems need positive, authoritative, and structured evidence to advance a lender from a mention to a shortlist recommendation. Neutral appearances do not satisfy that threshold.

The concentration pattern is self-reinforcing. Lenders that earn recommendation credit appear in more AI responses, which generates more public source evidence, which strengthens recommendation eligibility further. Lenders outside the top tier face a widening gap that requires deliberate investment in their AI discovery architecture to close.

The Category's Most Visible Warning Sign

Own Up appears in 30 AI observations across three clusters and receives zero recommendation credit on every platform tested. It is mentioned neutrally in discovery, comparison, and pricing prompts, but no AI system has ever recommended it as a lender option. This is the clearest example in the category of a brand that is visible but commercially irrelevant in AI-driven borrower discovery.

The warning sign is not that Own Up is unknown. It is that being known to AI systems is producing no shortlist placement. Neutral mentions in AI responses do not move borrowers toward a decision. They may actually reinforce competitor recommendations by providing context around lenders that do earn positive credit. Own Up is appearing in the same responses as Rocket Mortgage without receiving the same treatment. That asymmetry is the real exposure.

What This Means for the Category

The mortgage rates category is experiencing rapid shortlist compression. One lender captures more than 70% of the AI recommendation opportunity, and the remaining lenders are competing for a shrinking share of borrower attention. This concentration will intensify as AI platforms refine their recommendation outputs and borrowers become more reliant on AI-generated shortlists at the start of their mortgage search.

Lenders outside the top tier face a structural disadvantage that cannot be addressed through brand spend alone. Being visible in AI responses is not sufficient. Lenders need the entity, content, source, and citation architecture that allows AI systems to retrieve, trust, and advance their brand across multiple buyer stages and platforms. Without that foundation, visibility continues to compound for competitors rather than for themselves.

Platform performance varies significantly across the category. A lender that performs well on Copilot may be absent from Gemini. A lender that leads on Perplexity may be invisible on Google AI Mode. Competing for AI discovery requires a platform-specific strategy built on an understanding of where each lender is earning recommendation credit and where it is not.

AI discovery is becoming a permanent and growing part of the mortgage shopping journey. The lenders that invest in their AI recommendation architecture now are building a compounding advantage. The lenders that do not are deferring a structural problem that grows harder to close with each month of concentrated shortlist formation.

What This Public Benchmark Does Not Include

- Full cluster dataset across all 10 buyer stages

- Prompt-level response tables showing exact AI outputs

- Citation-source failure maps identifying missing evidence layers

- Platform-by-platform recovery priorities for each lender

- Entity and schema diagnostics for AI discoverability

- Source-layer gap analysis for recommendation eligibility

- Company-specific content recommendations

- Exact competitor threat profiles by platform and cluster

- Full paid opportunity model across all clusters

This page shows the market shape. The paid report shows the repair map.

Methodology and Disclaimers

1. Market studied: Mortgage Rates, including lender discovery, comparison, and pricing evaluation.

2. Brands and entities included: Rocket Mortgage, Better Mortgage, New American Funding, loanDepot, NerdWallet, LendingTree, Rate, AmeriSave Mortgage, and Own Up. The universe may not include all lenders active in the market.

3. Data collection window: May 2026.

4. AI platforms tested: ChatGPT, Copilot, Gemini, Perplexity, Google AI Mode, and Google AI Overviews.

5. Observations analyzed: 887 observations across three public clusters. Prompt count was not separately reported.

6. Prompt categories: Discovery and ranking prompts, head-to-head comparison prompts, and cost and pricing evaluation prompts.

7. Definition of a mention: A mention means the company appeared in an AI-generated response, regardless of sentiment or recommendation status.

8. Definition of a valid recommendation: A valid recommendation is a positive, shortlist-quality or ranked recommendation that earns recommendation credit. Visibility is not the same as recommendation credit.

9. Metrics used: Valid recommendation coverage, top-three rate, rank-one rate, average recommended rank, net sentiment score, and modeled monthly captured recommendation value.

10. Limitations: This is a point-in-time benchmark. AI outputs can change without notice. Modeled values are estimates and do not represent revenue. This report is not a full audit or complete market census.

For a Company-Specific Authority Index Report

For a company-specific Authority Index report, the deeper analysis would show which prompts each company wins or loses, which AI platforms are under-recognizing the brand, which source layers are shaping recommendations, and what changes may improve AI shortlist eligibility.