How AI Search Is Recommending Best Banks: Monthly Trends

Executive Summary

Ally Bank takes the recommendation lead in October 2026 at 83.1% valid recommendation coverage, ahead of Capital One in second at 82.7%. The gap is 0.4 points. That is a decisive change from September 2026, when Chase led at 24.4% and Ally Bank sat third at 22.1%. The category leaderboard was rebuilt this month, not merely reshuffled.

The strongest upward mover is Capital One, which reached 82.7% coverage in October 2026 from a 0.0% baseline reading in July 2026, a rise of 82.7 points. Its top-three placement reached 68.2% and its rank-one placement 19.2%. Ally Bank rose 47.5 points from 35.6% in July 2026 to 83.1% in October 2026, its single largest monthly move being a 61.0-point gain from September. Marcus by Goldman Sachs rose 42.4 points over the same baseline-to-current span, from 13.1% to 55.5%.

The sharpest decliner is Bank of America, down 29.5 points from 29.5% in July 2026 to 0.0% in October 2026, with a three-month declining streak. Chase, U.S. Bank, Wells Fargo, and Capital One Auto Finance each also register at 0.0% coverage in October 2026 against their July 2026 baselines of 34.3%, 23.1%, 27.3%, and 5.8% respectively. This coincides with a change in how tracked brand entities were named in October, which the benchmark discloses rather than folds into the trend line.

The October 2026 result sits at the end of a four-month series that ran 671, 679, 669, and 620 qualified observations. The category pattern is mixed: six brands classified as significant risers, six as significant decliners, and one stable.

Each monthly run begins with 800 prompt-surface observations (639 unique questions in October 2026) across the benchmark's defined AI/search surface universe. All 800 prompts mentioned a tracked brand or competitor; 751 were relevant and 49 were irrelevant. The public metrics use the 620 observations that survived both qualification stages. For comparison, July 2026 began with 800 prompt-surface observations (660 unique questions), 800 brand mentions, 707 relevant and 93 irrelevant, against 671 qualified observations.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• Which savings APY and fee figures did AI platforms attribute differently to Ally Bank and Capital One?
• Where did platforms disagree about Ramit Sethi's personal savings account recommendation?

Six critical or high-severity factual inconsistencies were detected this month across six AI platforms, concentrated in two companies.

Ally Bank

Savings APY (pricing, high severity). When asked "Who is the best online banking?", Google AI Mode stated Ally Bank offers 3.00% APY on savings, citing Bankrate's "Best Online Banks Of 2026," NerdWallet's "Best Online Banks," and SoFi's banking page. Copilot stated 4.20% savings APY for the same question, citing Bankrate, U.S. News, and The Wall Street Journal's "Best Online Banks and Rates - August 2026." The flagged sources included a third-party comparison page asserting a 4.20% online bank rate tracked in September 2026, a ranked-list page showing "Ally Checking+Savings 4.20%," and a CNBC Select page that applied a 4.20% APY figure to a different institution entirely.

Savings APY (pricing, high severity). When asked "Who has the best internet banking?", Copilot stated a 4.10% savings APY, citing Tradingpedia, Bankrate, and U.S. News. Google AI Mode stated 3.00% APY on savings for the same question, citing Bankrate, NerdWallet, and SoFi. A flagged source, a ranked third-party list, attributed "Ally Bank 4.10%" directly to the brand.

Ramit Sethi's personal savings account recommendation (factual, high severity). When asked "What savings account does Ramit Sethi recommend?", ChatGPT stated that Ally is only an alternative or secondary option rather than his personal pick, citing a book excerpt and interview transcript. Copilot stated that Ally Bank is his top recommendation and personal choice, citing the author's own site. One flagged source on the ChatGPT side quoted the author saying "Capital One 360 / ING Direct (This is the one I use)," while two flagged sources on the Copilot side quoted "1. Ally Bank (My personal choice)."

Monthly maintenance fee (pricing, high severity). When asked "3 apy savings account", Google AI Overviews stated that Ally Bank Advantage Savings carries an $8 monthly maintenance fee that can often be waived, citing an APY calculator page and Ally's own savings product page. Google AI Mode stated that Ally Bank Online Savings has no monthly maintenance fees, citing the WSJ, Investopedia, and a Google-hosted reference. Multiple flagged sources supported the no-fee claim, including Ally's own product page: "Along with a competitive, variable rate and no monthly maintenance fees."

Capital One

Savings account APY (pricing, high severity). When asked "Is Capital One savings a good idea?", Google AI Mode stated that Capital One 360 Performance Savings currently offers 3.10% APY, citing NerdWallet's review, a third-party review, and Capital One's own product page. Copilot stated the account offers around 4.00 to 4.10% APY in 2026, citing Money Atlas, Wealthvieuw, and ClearValueLending. A flagged source supported the 3.10% figure with a September 29, 2026 Bankrate reference and a caution to confirm the current rate; a flagged Money Atlas excerpt supported 4.00% APY with "no tiers, no conditions."

Ramit Sethi's personal savings account recommendation (factual, high severity). When asked "What savings account does Ramit Sethi recommend?", ChatGPT stated Capital One 360 is his personal go-to and the closest match to his personal setup, citing a book excerpt and interview transcript. Gemini stated that Ally Bank is his primary personal choice and go-to recommendation, citing the author's own site pages. A flagged source on the Gemini side quoted "1. Ally Bank (My personal choice) Ally has been my go-to savings account for years."

Benchmark Context

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set.

The qualified surface breadth held at six AI and search families in both months, even as question volume and qualification outcomes shifted underneath it. The table below turns from collection-stage counts to the benchmark-level recommendation metrics that drive this report's brand comparisons.

Benchmark-Level Metrics

August 2026 sat at 679 qualified observations with a 28.6% recommendation-shaped answer share and a 41.7% valid recommendation shortlist share; September 2026 sat at 669, 25.4%, and 34.7%. The three months that preceded October saw a shrinking shortlist pool. October reverses that: roughly two-thirds of qualified observations now carry a recommendation-shaped answer, and 85.3% carry a valid recommendation shortlist. That is a materially larger pool in which any brand can earn credit, and it is the single most important context for reading this month's coverage figures.

AI Recommendation Trend

Questions This Section Answers

• How tight is the gap between the top two banks in October 2026, and who leads the category?
• Which banks register at 0.0% coverage against their July 2026 baselines?

The Category Leaderboard Resets Around a Tight Two-Brand Top Tier

Ally Bank and Capital One now sit within 0.4 points of each other at the top, with Marcus by Goldman Sachs 27.2 points behind in third. Four brands register at 0.0% coverage against their July baselines.

Six brands exceeded the category's normal month-to-month variation on the upside and six on the downside. Only Chase Credit Journey sat inside it. The category-level change came from these movements in combination rather than from any single brand's shift, and the split between the entity names tracked in October and those tracked in the July-to-September run accounts for a substantial share of the downside column.

What Changed This Month

Questions This Section Answers

• Which banks were the largest risers and decliners in October 2026, and by how much?
• How does the October entity naming change affect the 0.0% readings for Chase, U.S. Bank, and Wells Fargo?

Ally Bank (category leader, significant riser)

Ally Bank's valid recommendation coverage moved from 35.6% in July 2026 to 83.1% in October 2026, a rise of 47.5 points. The prior-to-current move is sharper still: 61.0 points up from 22.1% in September 2026, the largest single-month move in the series. That is a significant riser on both comparisons.

Placement moved with coverage. Top-three placement reached 71.8% in October 2026, against 25.6% in July 2026, a gain of 46.2 points. Rank-one placement reached 27.7%, against 14.8% in July 2026, a gain of 12.9 points. The absolute counts are large: 445 top-three placements and 172 rank-one placements from 515 valid recommendations.

Raw mention presence rose from 66.0% in July 2026 to 90.3% in October 2026, a 24.3-point gain (443 to 560 observations). Net sentiment also improved, from about 0.7 to about 0.9. This is the rare case where presence, placement, and sentiment all move in the same direction.

Highest-priority diagnostic: Which prompts and surfaces converted from presence-only to active recommendation, and does the conflicting APY and fee information surfaced across platforms sit inside those same prompts?

Capital One (largest baseline-to-current mover)

Capital One's valid recommendation coverage reached 82.7% in October 2026 from a 0.0% July 2026 baseline, an 82.7-point rise. There is no September comparison row for Capital One because the entity was not tracked under that name in the earlier months.

Presence and placement both landed high. Raw mention presence reached 98.4%, the highest in the category, at 610 of 620 observations. Top-three placement reached 68.2% (423 placements) and rank-one placement 19.2% (119 placements). Valid recommendations totalled 513.

The distinction to notice is that Capital One leads the category on raw presence but sits second on recommendation coverage, 0.4 points behind Ally Bank. Being named most often and being recommended most often are separate signals, and here they point at different brands.

Highest-priority diagnostic: Which prompts surface Capital One without recommending it, and do the conflicting APY figures returned across platforms concentrate in those prompts?

Marcus by Goldman Sachs (significant riser)

Marcus by Goldman Sachs moved from 13.1% coverage in July 2026 to 55.5% in October 2026, a rise of 42.4 points. The prior-to-current move is the largest in the category, 48.9 points up from 6.6% in September 2026.

The gain is supported at every level. Raw mention presence rose 38.4 points, from 24.0% to 62.4% (161 to 387 observations). Top-three placement rose 25.0 points, from 8.2% to 33.2%, and rank-one placement rose 5.7 points, from 1.6% to 7.3%. Net sentiment moved from about 0.7 to about 0.9. This is a broad gain rather than a single-axis shift.

Highest-priority diagnostic: Which surfaces drove the 38.4-point presence gain, and is the brand now appearing in prompts where it previously received no mention at all?

Bank of America (significant decliner, three-month streak)

Bank of America's valid recommendation coverage moved from 29.5% in July 2026 to 0.0% in October 2026, a decline of 29.5 points. The prior-to-current move was 22.4 points down from 22.4% in September 2026, also significant, giving the brand a three-month declining streak.

A related entity, Bank of America Corp., was tracked separately in October 2026 and registered 4.0% coverage, up 4.0 points from a 0.0% baseline, with raw mention presence of 24.5% (152 observations) and top-three placement of 2.1%.

The distinction the reader should carry: the aggregate trend line for Bank of America reflects the entity name that was tracked through September. The October run introduces a different entity name. The two are not directly continuous, and the 0.0% reading should be read as a tracking change rather than as a disappearance of the underlying institution from AI answers.

Highest-priority diagnostic: How do AI systems refer to this institution across surfaces in October 2026, and how much of the July-to-September trend would persist if the entity naming were held constant?

Chase, U.S. Bank, and Wells Fargo (significant decliners, three-month streaks)

Chase moved from 34.3% coverage in July 2026 to 0.0% in October 2026, down 34.3 points, with a 24.4-point prior-to-current decline from 24.4% in September 2026. U.S. Bank moved from 23.1% to 0.0%, down 23.1 points, with an 18.7-point prior-to-current decline. Wells Fargo moved from 27.3% to 0.0%, down 27.3 points, with a 20.8-point prior-to-current decline. All three carry three-month declining streaks.

Chase Credit Journey, a distinct entity, registered 0.2% coverage with 0.5% raw mention presence and 3 observations, classified stable. U.S. Bancorp registered 1.9% coverage from a 0.0% baseline with 16.0% raw presence and 99 observations. Wells Fargo & Co. registered 2.4% coverage from a 0.0% baseline with 20.0% raw presence and 124 observations.

The pattern across these three pairs is the same: the entity names tracked through September now read 0.0%, while closely related entity names entered the October run at low single-digit coverage on modest observation counts. The reader should treat the 0.0% readings as a naming continuity question, not as evidence that these institutions vanished from AI answers.

Highest-priority diagnostic: Which surface or surfaces changed how these institutions are named, and does the low October coverage for the new entity names come from the same prompt set the earlier names served?

Discover Home Loans (significant decliner, small counts)

Discover Home Loans moved from 3.9% coverage in July 2026 to 0.2% in October 2026, a decline of 3.7 points, with a 1.4-point prior-to-current decline from 1.6% in September 2026. Both moves are significant, and the brand carries a two-month declining streak.

The absolute counts are tiny. One valid recommendation in October 2026. Raw mention presence moved from 7.8% in July 2026 to 0.2% in October 2026, a 7.6-point decline (52 to 1 observation). Top-three placement fell from 0.8% to 0.0%.

This is a small-count brand and the movement sits on a single observation. That single positive mention happened to carry a net sentiment score of 1.0, which is worth noting without over-reading.

Highest-priority diagnostic: Which single prompt produced the one remaining mention, and is the brand consistently available to AI systems at all under its current naming?

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters do the qualified observations cover, and which remain unmeasured?
• Why can't the public benchmark separate rate positioning or head-to-head comparison outcomes?

All 620 qualified observations in October 2026 fell into the brand recommendation class. No qualified observations were classified into the pricing and value or multi-brand comparison clusters. The public benchmark can therefore speak to which bank AI systems recommend for a general best-bank query, but it cannot yet separate outcomes for rate positioning, fee competitiveness, or head-to-head comparison prompts. The inconsistency data above shows why that gap matters commercially: when the same savings product is attributed a 3.00% APY by one platform and a 4.20% APY by another, the reader has evidence of a divergence the public coverage metric cannot quantify.

Brand Opportunity Summary

Questions This Section Answers

• What is the highest-priority diagnostic for each bank's October 2026 coverage signal?
• Which banks lead, which are risers, and which carry declining streaks?

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations (query, surface, recommendation outcome, rank, sentiment, and citations where exposed). Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns. Source presence is not automatically treated as proof of causation.

About This Benchmark

This report is part of the LLM Authority Index AI Visibility Market Discovery research program.

• AI Visibility Industry Market Methodology
• AI Visibility Industry Market Metrics
• AI Visibility Industry Market Standards

Report-Specific Interpretation Notes

• Small-count brands (Chase Credit Journey, Discover Home Loans, U.S. Bancorp, Wells Fargo & Co., Bank of America Corp.) carry wider variation risk; their absolute counts are named beside percentages so movement can be read in context.
• The qualified denominator (620) differs from the raw collection (800); all percentages are calculated against the qualified set.
• Movement between months identifies changes worth investigating; it does not by itself establish the cause of those changes.
• The October run introduces a larger valid recommendation shortlist share than the three preceding months. Part of this month's coverage movement reflects a wider pool in which brands can earn credit.
• Entity naming changed between the July-to-September run and October. Four entity names present through September read 0.0% in October while related October entity names carry low single-digit coverage; those pairs are not directly continuous.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

The aggregate percentage hides the decisions driving it. Which high-intent prompts are won? When a brand loses, which competitor takes the recommendation? What attributes do AI systems associate with each bank, and which external sources shape those answers? When two platforms return a 3.00% APY and a 4.20% APY for the same savings product, the public benchmark records the recommendation but cannot explain the source of the disagreement. A public benchmark cannot answer those questions.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. It turns the movement you see here into a plan.

Request an AI visibility audit