How AI Search Is Recommending Password Managers: Monthly Trends

Executive Summary

Bitwarden remains the category leader in October 2026, holding valid recommendation coverage of 89.2%, up 8.4 points from 80.8% in July. The gain makes Bitwarden one of two significant risers in the series this period. The gap to second-place 1Password, which sits at 80.5% coverage, is 8.7 points, up from 2.7 points in July. Bitwarden's move is the largest baseline-to-current gain in the category.

1Password holds second place with coverage of 80.5%, up 2.4 points from July's 78.1%, a change within normal variation. The brand remains the rank-one leader, holding top placement in 48.8% of qualified observations, up 12.8 points from 36.0% in July. Its top-three rate rose 8.6 points to 78.8% from 70.2%, even as raw mention presence fell 3.6 points to 90.9%.

NordPass is the category's other significant riser, with coverage of 67.8%, up 6.5 points from 61.3% in July. Its top-three rate rose 6.3 points to 42.7%, and its raw mention presence climbed 5.5 points to 75.3%. No tracked brand registered a significant decline in valid recommendation coverage across the baseline period.

Against the prior month, 1Password gained 4.7 points from September's 75.8%, and Bitwarden gained 5.1 points from 84.1%, a significant prior-month move. RoboForm's coverage fell 3.6 points from September's 34.5%, and Proton Pass fell 1.5 points from 62.5%, its second consecutive monthly decline. Across the full July-to-October series, the two-brand leadership structure held while the upper mid-tier expanded its coverage.

Each monthly benchmark run begins with 800 prompt-surface observations across the benchmark's defined AI/search surface universe: 462 unique questions in October and 390 in July, with all 800 prompts in both months mentioning a tracked brand or competitor. Of those, 778 were relevant and 22 irrelevant in October, versus 784 relevant and 16 irrelevant in July. The public metrics use the 518 qualified observations in October and 530 in July that survive both qualification stages.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• How many factual inconsistencies did the AI platforms surface, and what topics did they cover?
• Which brand drew the highest concentration of conflicting claims?
• Where did platforms disagree most sharply on breach history and free tier availability?

The benchmark detected 14 critical or high-severity factual inconsistencies across six AI platforms: ChatGPT, Copilot, Gemini, AI Mode, AI Overviews, and Perplexity. The inconsistencies span six companies and cover free tier availability, breach history, pricing, and product feature claims.

Dashlane

Dashlane accounts for seven of the fourteen detected inconsistencies, the highest concentration in the category. Four separate conflicts concern free tier availability. When asked "Is Dashlane a good password manager?", Google AI Overviews stated that the free version "is heavily restricted to just 50 passwords," while Google AI Mode stated that "Dashlane discontinued its free tier, meaning you have to pay for a subscription to use it across multiple devices meaningfully." ChatGPT stated "the free version exists but is quite restricted compared with some alternatives," while Gemini stated that "it does not offer a permanent free tier anymore, requiring a paid subscription (or free trial) to use it." The conflicting answers cited PCMag, Security.org, All About Cookies, and PasswordManager.com, with flagged excerpts showing source pages disagreeing on whether the free plan allows 25 passwords, 50 passwords, or no permanent tier at all.

A separate high-severity conflict concerns breach history. When asked "How trustworthy is Dashlane?", ChatGPT stated that Dashlane "had a May–June 2026 brute-force attack where fewer than 20 personal-plan users had encrypted vaults downloaded, plus a 2025 third-party CRM incident and a 2025 cryptographic downgrade vulnerability," referencing Dashlane's own architecture overview and a USENIX presentation. Google AI Mode stated that Dashlane "has maintained a clean record with zero known company-wide security breaches compromising user vaults since its inception," citing Dashlane's compliance certifications page and an All About Cookies review. A related conflict appeared when the question changed to "Which password manager has been hacked?": ChatGPT again described the 2026 encrypted-vault incident, while Gemini stated that Dashlane "has not experienced a direct server breach where encrypted user vaults were stolen." A flagged excerpt from Dashlane's own security advisory page aligns with the ChatGPT response.

NordPass

NordPass accounts for three detected inconsistencies across five platforms. When asked "Is there a 100% free password manager?", Google AI Mode stated that "NordPass free tier restricts you to a single device type or severely limits the number of passwords you can save," while Perplexity stated that "NordPass is a strong free option with unlimited passwords (some features may be restricted)." Flagged source excerpts disagreed in the same way, with one excerpt stating "free plan includes all basic features and supports unlimited passwords" and another stating "unlimited password storage ... can only be logged in on one device at a time."

A second conflict concerns built-in TOTP authenticator availability. When asked "What are the downsides of NordPass?", ChatGPT stated that "NordPass's current documentation advertises a built-in Authenticator/TOTP feature, so the historical lack of TOTP has been addressed," citing NordPass's own premium-versus-free support page and a WIRED review. Gemini stated that "NordPass lacks built-in authenticator code generation for stored accounts." The flagged WIRED excerpt sided with Gemini, while excerpts from NordPass support and PanicVault sided with ChatGPT.

A third conflict concerns free tier device limits. When asked "What is the best password manager software?", Google AI Mode stated NordPass allows "1 device at a time," while Copilot stated "unlimited devices," with one flagged retail listing describing a "NordPass Premium, Unlimited Devices" subscription.

1Password

One inconsistency was detected. When asked "What is the best password manager for 2025?", Copilot stated a monthly price of "~$24.95/mo (premium only)," while Google AI Overviews stated that "personal plans start around $48/year." The two figures describe incompatible annualized costs for the same personal plan, and the conflicting answers cited The Independent, How-To Geek, PCMag, and WIRED.

Bitwarden

One inconsistency was detected. When asked "Which password manager has never been hacked?", Perplexity listed Bitwarden "among password managers that have experienced breaches," while Google AI Overviews stated that "Bitwarden has never suffered a security breach or been hacked." A flagged excerpt sided with the Google AI Overviews response.

LastPass

One inconsistency was detected. When asked "What is the best password manager software?", Copilot stated that "LastPass is a solid choice for simplicity and affordability," while Google AI Mode stated that "experts generally advise avoiding LastPass due to its history of major data breaches." The flagged excerpts split on the same question.

Proton Pass

One inconsistency was detected. When asked "Is Proton Pass truly free to use?", ChatGPT stated that the "free tier includes a built-in 2FA authenticator," while Copilot stated that "integrated 2FA requires a paid upgrade; free users are capped at only 3 two-factor authentication items." The flagged excerpts sided with the Copilot framing.

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 funnel narrowed steadily across the series: 538 qualified observations in August and 528 in September preceded October's 518, with irrelevant prompts rising from 16 in July to 33 in September before settling at 22 in October.

Benchmark-Level Metrics

The recommendation-shaped answer share peaked at 65.6% in August, settled to 51.5% in September, then climbed to 67.0% in October, the highest reading in the series. Valid recommendation shortlist share followed a similar pattern, dipping to 84.5% in September before reaching 90.3% in October.

AI Recommendation Trend

Questions This Section Answers

• Which password managers registered significant coverage gains against the July baseline?
• How did the gap between the two leaders and the rest of the field change?
• Which brands carried the only coverage declines in the series?

The two-brand leadership structure held, with Bitwarden ahead of 1Password, while the upper mid-tier expanded and the gap between the leaders and the rest of the field widened.

Bitwarden and NordPass each registered a significant gain against the July baseline on the primary coverage metric. The remaining category-level change came from a combination of smaller movements, with the two leaders and NordPass accounting for the widest gains and RoboForm and LastPass carrying the only declines.

What Changed This Month

Questions This Section Answers

• Which brands moved most in October, and how do their coverage and placement shifts differ?
• Where did raw mention presence diverge from rank-one or top-three placement?

Bitwarden

Bitwarden's valid recommendation coverage reached 89.2% in October, up 8.4 points from 80.8% in July, a significant baseline gain. The move also represents a 5.1-point gain over September's 84.1%, itself a significant month-over-month change; September's figure had pulled back from August's 89.0% peak.

The brand's top-three recommendation rate rose 11.9 points to 81.3% from 69.4% in July. Its rank-one rate moved up 0.2 points to 31.1% from 30.9%, little changed. Raw mention presence climbed 2.4 points to 98.8% from 96.4%, leaving only 6 of 518 qualified observations in October that did not mention the brand.

The distinction: Bitwarden's October recovery is broader than its coverage number alone suggests. The brand returned to, and slightly exceeded, its August peak level on coverage while expanding its top-three rate by nearly twelve points against the July baseline. Its presence remains close to saturation.

Highest-priority diagnostic: Which surfaces drove the September-to-October recovery, and whether the top-three expansion came from new prompt categories or deeper penetration in existing ones.

1Password

1Password's valid recommendation coverage reached 80.5% in October, up 2.4 points from 78.1% in July, a change within normal variation. That figure also represents a 4.7-point recovery from September's 75.8%.

The brand's rank-one rate rose 12.8 points to 48.8% from 36.0% in July, keeping 1Password as the clear top-placement leader. Its top-three rate rose 8.6 points to 78.8% from 70.2%. Raw mention presence fell 3.6 points to 90.9% from 94.5%, meaning the brand was absent from roughly 47 of 518 qualified observations in October.

The distinction: 1Password is appearing in fewer answers while winning the top slot more often when it appears. The divergence between its presence decline and its rank-one gain suggests the brand's strength is concentrated in the answers where it is placed highest, not in the breadth of its presence.

Highest-priority diagnostic: Which answer types dropped 1Password from their shortlists, and whether the rank-one gain is concentrated on specific AI surfaces.

NordPass

NordPass's valid recommendation coverage reached 67.8% in October, up 6.5 points from 61.3% in July, a significant baseline gain. The move follows a 5.5-point prior-month decline from August's 71.0% to September's 65.5%.

The brand's top-three rate rose 6.3 points to 42.7% from 36.4% in July. Raw mention presence climbed 5.5 points to 75.3% from 69.8%. Rank-one placement moved down 0.9 points to 5.0% from 5.9%, little changed.

The distinction: NordPass is expanding both its presence and its placement quality faster than its baseline coverage gain alone indicates. The brand holds the widest gap to the leader among the top three, but its gains on presence and top-three placement separate it from the stable brands beneath it.

Highest-priority diagnostic: Which surfaces account for the presence gain, and whether the top-three improvement is concentrated in recommendation-shortlist answers or spread across answer types.

Proton Pass

Proton Pass's valid recommendation coverage reached 61.0% in October, up 2.7 points from 58.3% in July. The brand's coverage declined for a second consecutive month, down 1.5 points from September's 62.5% and from August's 65.8% peak, though neither monthly move crossed the significance threshold for the primary metric.

The brand's top-three rate rose 0.1 points to 32.6% from 32.5% in July. Raw mention presence fell 0.4 points to 64.5% from 64.9%. Rank-one placement moved down 0.6 points to 1.7% from 2.3%.

The distinction: Proton Pass holds a stable upper-tier position with zero negative mentions in October, but its coverage has drifted down from the August peak across two consecutive months. The brand's net sentiment score of 1.0 is the highest in the category.

Highest-priority diagnostic: Where the August gains consolidated, and which prompts drove the September-to-October pullback.

RoboForm

RoboForm's valid recommendation coverage fell to 30.9% in October, down 1.7 points from 32.6% in July and down 3.6 points from September's 34.5%. Neither movement crossed the significance threshold for the primary metric.

The brand's top-three rate rose 6.5 points to 12.0% from 5.5% in July. Its rank-one rate moved down 0.2 points to 2.1% from 2.3%. Raw mention presence fell 4.1 points to 35.1% from 39.2%. The brand recorded 160 valid recommendations in October, down from 173 in July.

The distinction: RoboForm's placement quality improved even as its coverage and presence declined. The brand is being recommended in fewer answers but placed higher within the answers that include it. This is a small-count effect on the rank-one measure, with 11 observations in October.

Highest-priority diagnostic: Which prompt types drove the top-three improvement, and whether the presence decline is concentrated on surfaces where the brand previously held a stable position.

Keeper

Keeper's valid recommendation coverage reached 35.7% in October, up 2.5 points from 33.2% in July and up 3.5 points from September's 32.2%, both within normal variation for the primary metric.

The brand's top-three rate rose 7.1 points to 14.3% from 7.2% in July. Its rank-one rate rose 1.5 points to 1.7% from 0.2%. Raw mention presence moved up 1.9 points to 43.6% from 41.7%. The brand recorded 185 valid recommendations in October, up from 176 in July.

The distinction: Keeper's placement quality improved on both top-three and rank-one measures, but its coverage gain remained inside normal variation. The brand's coverage now sits above its September trough but below its July-to-September band, indicating the improvement is more about placement than breadth.

Highest-priority diagnostic: Which prompts drove the top-three and rank-one gains, and whether they cluster on specific AI surfaces.

LastPass

LastPass's valid recommendation coverage fell to 3.7% in October, down 1.2 points from 4.9% in July, within normal variation. The move also represents a 0.9-point recovery from September's 2.8% trough. The brand recorded 19 valid recommendations in October, up from 15 in September but down from 26 in July.

Raw mention presence fell 17.5 points to 23.8% from 41.3% in July, the sharpest decline in the category. The brand's net sentiment score improved to -0.1 from -0.3 in July. It carried 41 negative mentions in October against 23 positive mentions, down from 98 negative mentions in July.

The distinction: LastPass's presence decline spans the full baseline period even as its coverage recovered slightly from September's trough. The brand is mentioned in fewer than a quarter of qualified answers, and its negative-to-positive mention ratio remains inverted.

Highest-priority diagnostic: Which prompt categories have dropped LastPass from their answers entirely, and whether the remaining mentions cluster in factual or cautionary responses rather than recommendation shortlists.

Dashlane and Enpass

Dashlane's valid recommendation coverage reached 28.0% in October, down 0.5 points from 28.5% in July and up 0.9 points from September's 27.1%, both within normal variation. Its top-three rate fell 0.4 points to 6.6% from 7.0%, and its rank-one rate fell 0.4 points to 0.2% from 0.6%, holding a single rank-one observation in October. Raw mention presence fell 1.6 points to 42.7% from 44.3%.

Enpass's coverage held at 0.6% in October, flat against July. The brand recorded 3 valid recommendations in October, unchanged from July. Raw mention presence fell 1.3 points to 1.0% from 2.3%, and net sentiment rose to 0.6 from 0.2.

The distinction: both brands sit near their July baseline with no significant movement in either direction. Dashlane's rank-one position, at a single observation, is a small-count measure that does not support directional conclusions. Enpass's three valid recommendations place any coverage change within a fraction of a percentage point.

Highest-priority diagnostic: Which small set of prompts still recommends Enpass, and whether Dashlane's rank-one presence is concentrated on a single surface.

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters did the qualified observations actually cover?
• What can the benchmark not yet answer because of where the qualified observations fell?

The October qualified observations fell entirely within the Brand Recommendation cluster (518 observations), with zero qualified observations in the Multi-Brand Comparison or Pricing & Value clusters. The same pattern held across every month in the series.

The public benchmark therefore cannot yet answer price, value, or head-to-head comparison questions for the password managers vertical. The competitive signals in this report reflect recommendation behavior on discovery-style prompts only. The pricing and feature conflicts surfaced in the inconsistency analysis above show that this buyer-intent gap matters: AI platforms are answering pricing and free tier questions inconsistently even though those questions sit outside the qualified benchmark set.

Brand Opportunity Summary

Questions This Section Answers

• Which password managers show the strongest current signals, and which need diagnostic attention?
• What highest-priority diagnostic does each brand's position point to?

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 CiteWorks Studio AI Visibility Market Discovery research program.

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

Report-Specific Interpretation Notes

• Small-count brands (Enpass with 3 valid recommendations, LastPass with 19, Dashlane with 1 rank-one observation) show movements that may not persist; absolute counts are named beside percentages.
• The qualified denominator (518 observations) differs from the raw collection (800 prompts); brand-level percentages reflect only qualified observations that survived both qualification stages.
• Movement between months is directional analysis that identifies changes worth investigating, not proof of cause.
• The July-to-October series uses July as baseline; August and September figures belong in prose, not table columns.

Next Step

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

Beneath the aggregate coverage percentage sit questions the public benchmark cannot answer: which high-intent prompts a brand wins or loses, which competitor takes the recommendation when a brand loses the slot, and which external sources shape those answers. The inconsistency analysis above shows that AI platforms are surfacing conflicting claims about free tier availability, breach history, pricing, and feature sets, and those conflicts sit upstream of the recommendation behavior the benchmark measures.

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 benchmark's directional signals into actionable intelligence for a specific brand's position.

Request an AI visibility audit