How AI Search Is Recommending Car Insurance: Monthly Trends

Executive Summary

Mercury Insurance remains the category leader in AI-driven car insurance recommendations with 30.9% valid recommendation coverage in October 2026, ahead of Root Insurance at 24.5%, Direct Auto Insurance at 18.1%, and a tie at 14.2% between Clearcover and Mile Auto. Mercury's lead over the second-place brand widened to 6.4 points this month, and its coverage now sits 28.8 points above its 2.1% May 2026 baseline, the largest baseline-to-current gain in the category.

Mercury Insurance is again the strongest upward mover with any meaningful scale, up 5.8 points from 25.1% in September to 30.9% in October, its second consecutive monthly gain. Its top-three rate reached 14.9% and its rank-one rate reached 4.3% with 12 rank-one placements. Mile Auto posted the sharpest prior-to-current decline among established brands, down 4.8 points from 19.0% in September to 14.2% in October with 40 valid recommendations, though it remains 14.2 points above its 0.0% baseline.

The most consequential movement in the October series is the case of The General. Across the full May-to-October window, The General recorded 8.6% coverage at the May baseline, rose as high as 27.9% in July, held 25.9% in August, then dropped to 0.0% in September and remained at 0.0% in October, a baseline-to-current decline recorded as significant. A separate tracked entity, The General®, registered 5.7% coverage in October after 5.0% in September, up 5.7 points from its 0.0% baseline.

This month did not produce a new category-level significant movement: no brand's month-over-month change was large enough to register as significant, even though several brands continue to move markedly against the May 2026 baseline. The category instead reflects the cumulative effect of a nine-brand cohort of significant risers against a May 2026 baseline that has now consolidated into a four-brand top tier.

Each monthly run begins with 800 prompt-surface observations (496 unique questions in October, 499 in September, 700 in August) across the benchmark's defined AI/search surface universe. Of those, 800 mentioned a tracked brand or competitor; 776 were relevant and 24 were irrelevant in October. The public metrics use the 282 qualified observations in October that survive both qualification stages, compared with 279 in September and 140 in the May baseline.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• Which companies had the most critical or high-severity AI response inconsistencies this month?
• Where did AI platforms contradict each other on car insurance pricing claims?
• What kinds of factual conflicts showed up most often across the six AI surface families?

The October benchmark detected 13 critical and high-severity factual inconsistencies across five companies and all six AI surface families (ChatGPT, Copilot, Gemini, Google AI Mode, Google AI Overviews, and Perplexity). Each conflict below reflects directly contradictory claims surfaced by two or more AI platforms in response to the same question.

Elephant Insurance

Six critical and high-severity inconsistencies were detected across five platforms for Elephant Insurance, all centred on current ownership.

In a critical conflict, AI platforms provided conflicting information about whether Admiral currently owns Elephant Insurance. When asked "Is Elephant owned by Admiral?" Perplexity stated that Elephant is owned by Admiral Group plc as a wholly-owned subsidiary, citing insuredandmore.com and the company's own press releases, while Gemini stated that Elephant is no longer owned by Admiral Group and was sold to J.C. Flowers & Co. effective December 31, 2025, citing Admiral Group's own news release confirming completion of the sale and Insurance Journal coverage. The flagged sources show the discrepancy originates in part from a legacy page still describing Elephant as a trading name of EUI Ltd, an Admiral Group company.

In a second critical conflict on the same question, Perplexity repeated the Admiral ownership claim while ChatGPT stated that Elephant Insurance is no longer owned by Admiral and was sold to J.C. Flowers & Co. effective December 31, 2025, citing Admiral Group's sale announcement and the Elephant company facts page. Flagged sources include the Elephant company facts page itself, which states the sale "will be completed on January 1, 2026," alongside a Wikipedia entry describing Admiral's 2009 entry into the US market with Elephant.

A high-severity conflict emerged when asked "Is Elephant a real insurance company?" Google AI Mode stated that Elephant was originally founded as a subsidiary of the UK-based Admiral Group and later acquired by the private equity firm J.C. Flowers & Co., while Google AI Overviews stated that it is a subsidiary of the massive UK-based Admiral Group plc. Flagged sources include the Elephant company facts page, which is being read both ways, and Business Insider's review describing Elephant as under the Admiral Group.

Two further high-severity conflicts on the same question repeated the same pattern: Google AI Mode's acquisition narrative against Perplexity's "subsidiary of the Admiral Group" claim, and against Gemini's "wholly owned subsidiary of the Admiral Group plc" claim. In the Gemini comparison, flagged sources include the Elephant info page stating that "in January 2026, Elephant joined the portfolio of companies owned by private equity firm J.C. Flowers & Co." alongside coveragecat.com stating the sale completed on January 1, 2026.

A sixth high-severity conflict was also detected when asked "Is Elephant a real insurance company?" where Google AI Mode described the J.C. Flowers acquisition while ChatGPT stated that Elephant was founded in 2009 and was previously part of Admiral Group plc, citing the company homepage and an Illinois Department of Financial and Professional Regulation filing.

Mercury Insurance

Three high-severity inconsistencies were detected across Copilot, Google AI Mode, and Google AI Overviews for Mercury Insurance.

In a high-severity availability conflict, platforms disagreed on the number of states where Mercury operates. When asked "Is Mercury Insurance in all 50 states?" Copilot stated that Mercury operates in 12 states including New Mexico, while Google AI Mode stated that Mercury operates in 11 states not including New Mexico. Flagged sources on the Copilot side include nationalinsuranceguide.com listing 12 states including NM, while the Google AI Mode side cited Mercury's own fact sheet and local pages.

In a high-severity pricing conflict, platforms disagreed on Mercury's price competitiveness. When asked "Is Mercury a good insurance company?" Copilot stated that Mercury's rates are significantly above average, at $270 per month against a national average of $209 and a six-month policy averaging $1,536, while Google AI Mode stated that Mercury offers highly competitive, affordable rates frequently landing in the cheapest quartile for drivers with clean records. Flagged sources on the "above average" side include insure.com and insurance.com, while the "competitive" side cited WalletHub and ChooseFI.

In a third high-severity pricing conflict, platforms disagreed on which insurer offers the cheapest minimum liability coverage in Arizona. When asked "Who has the cheapest car insurance for Arizona?" Google AI Overviews stated that Mercury is not listed among the cheapest, with Travelers averaging around $45 per month and USAA about $56 per month, while Copilot stated that Mercury Insurance is around $33 to $43 per month for minimum liability coverage, described as the cheapest option. Copilot's flagged sources include way.com and Stacker listing Mercury as the cheapest.

SafeAuto

Two high-severity inconsistencies were detected across Copilot and Google AI Mode for SafeAuto.

In a high-severity factual conflict, platforms disagreed on whether SafeAuto carries an AM Best financial strength rating. When asked "Is safe auto insurance legit?" Google AI Mode stated that AM Best assigns a Financial Strength Rating of A+ (Superior) as part of the Allstate group, citing a Business Wire release on AM Best credit ratings, while Copilot stated that SafeAuto was not rated by AM Best, raising questions about long-term stability. Copilot's flagged source was insuredbetter.com, which states the company was not rated by AM Best.

In a second high-severity factual conflict, platforms disagreed on SafeAuto's parent company. When asked "Is SafeAuto a good insurance company?" Google AI Mode stated that SafeAuto was acquired by Allstate and integrated into Direct Auto Insurance, while Copilot stated that SafeAuto is now part of Kemper and described it as absorbed into Direct Auto Insurance under Allstate's National General subsidiary. This conflict combines an ownership claim (Kemper versus Allstate) with a distribution claim (Direct Auto), both of which cannot be simultaneously correct.

Direct Auto Insurance

One high-severity inconsistency was detected across Gemini and Google AI Mode for Direct Auto Insurance.

In a high-severity pricing conflict, platforms gave incompatible averages for the same metric. When asked "What is the average direct auto quote?" Google AI Mode stated approximately $166 to $167 per month, while Gemini stated approximately $264 per month. Both platforms cited overlapping sources including autoinsurance.com and financebuzz.com, suggesting the divergence is not from clearly separate evidence bases.

Root Insurance

One high-severity inconsistency was detected across Copilot and Google AI Overviews for Root Insurance.

In a high-severity pricing conflict, platforms gave incompatible monthly full-coverage prices for Root. When asked "Who has the cheapest car insurance for Arizona?" Google AI Overviews stated that Root and Progressive have averages starting around $1,456 per year, or roughly $121 per month, for drivers with clean records, while Copilot stated that Root Insurance Company is about $79 to $80 per month for full coverage. Copilot's flagged sources include Stacker and way.com listing Root at $79 and $80 respectively.

Benchmark Context

Questions This Section Answers

• How does the qualified benchmark set differ from the raw prompt collection?
• What happened to qualified observations and unique questions between May and October 2026?

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

September 2026 reserved 489 prompts at the eligibility stage from 768 relevant prompts, while October reserved 494 from 776 relevant prompts. August 2026 sat between the two at 165 reserved from 489 relevant. The eligibility reservation has grown across the series, which is why qualified observations have settled into a tighter range in recent months even as the raw collection universe expanded to 800 prompts in July.

Benchmark-Level Metrics

October's valid recommendation shortlist share of 67.0% is the highest recorded in the series, exceeding July's 54.9% and August's 55.2%. The recommendation-shaped answer share (39.4%) is broadly consistent with September (40.1%) and August (46.0%).

AI Recommendation Trend

Questions This Section Answers

• Which car insurance brands made up the top tier in October 2026, and how far apart are they?
• How large were the month-over-month shifts among the leading brands?
• Which brands posted the largest gains from the May 2026 baseline?

The Category Consolidates Into a Four-Brand Top Tier

October reflects a stabilised leaderboard. Mercury Insurance holds first, Root Insurance second, Direct Auto Insurance third, and Clearcover and Mile Auto tie at 14.2%. The gap between the leader and the runner-up is now 6.4 points, up from 3.2 points in September 2026.

The month-over-month picture shows the category-level movement came from several smaller shifts rather than any single brand exceeding normal month-to-month variation. Mercury Insurance was the strongest upward mover, up 5.8 points, but the shift was not large enough to be recorded as significant at the category level. Mile Auto's 4.8-point decline was the sharpest pullback among established brands. Clearcover continued a four-month upward streak that began after its 5.8% July reading.

What Changed This Month

Questions This Section Answers

• Why did Mercury Insurance's lead over the runner-up widen in October?
• What drove Mile Auto's pullback from September to October?
• How should The General's 0.0% reading be read alongside The General®'s 5.7%?

Mercury Insurance: Widening Lead and Confirmed Rank-One Presence

Mercury Insurance reached 30.9% valid recommendation coverage in October, up 5.8 points from 25.1% in September and 28.8 points from its 2.1% May 2026 baseline. Mercury now leads the runner-up by 6.4 points, up from a 3.2-point gap in September.

Presence rose to 58.9% raw mention presence in October from 47.7% in September, with the brand appearing in 166 of 282 observations versus 133 of 279. Its top-three rate reached 14.9%, up from 11.1% in September, and its rank-one rate reached 4.3% with 12 rank-one placements.

Mercury leads the category in presence and coverage but still converts to rank-one placements at a relatively modest rate compared with Root Insurance's 6.7%, indicating broad visibility rather than first-choice dominance.

Highest-priority diagnostic: Which prompt categories and surface families concentrate Mercury's rank-one placements, and which competitor takes first position when Mercury is recommended second or third?

Root Insurance: Sustained Rank-One Strength

Root Insurance held second place at 24.5% valid recommendation coverage in October, up 2.6 points from 21.9% in September and 23.8 points from its 0.7% May 2026 baseline. Root's rank-one rate of 6.7% with 19 rank-one placements leads the top tier and now exceeds Mercury's 4.3%.

Presence held broadly similar to September at 32.3% raw mention presence, with the brand appearing in 91 of 282 observations versus 92 of 279. Its top-three rate reached 13.1%, and its net sentiment score registered 0.7, the third-highest among the top five, behind Mile Auto's 0.9 and Clearcover's 0.8.

Root converts its presence into top-three placements at a similar rate to Mercury but converts top-three placements into first position at a materially higher rate, making it the leader in rank-one density within the top tier.

Highest-priority diagnostic: Which evidence sources drive Root's rank-one placements versus its lower-position recommendations?

Mile Auto: Sharpest Pullback Among Established Brands

Mile Auto eased 4.8 points from 19.0% in September to 14.2% in October, the sharpest decline among brands with continuous tracking. Mile Auto remains 14.2 points above its 0.0% May 2026 baseline, and its valid recommendation count was 40 in October versus 53 in September.

Presence followed the decline, at 15.6% raw mention presence in October versus 22.2% in September, with the brand appearing in 44 of 282 observations versus 62 of 279. Its top-three rate eased to 8.2% from 11.1%, and its rank-one rate fell to 1.8% from 4.3%.

Mile Auto retains the highest net sentiment score among top-tier brands (0.9), meaning the pullback is in visibility volume rather than sentiment quality. The brand appears less often but continues to be favoured when it does appear.

Highest-priority diagnostic: Whether Mile Auto's October pullback is concentrated in Google AI Mode, where Mercury Insurance is the top authority company, or spread across multiple surfaces.

Clearcover: Four-Month Upward Streak

Clearcover reached 14.2% valid recommendation coverage in October, up 3.4 points from 10.8% in September and 14.2 points from its 0.0% May 2026 baseline. This is Clearcover's fourth consecutive monthly gain, the longest active upward streak in the category alongside Kemper Auto's four-month streak.

Presence rose to 17.7% raw mention presence in October from 14.0% in September, with the brand appearing in 50 of 282 observations versus 39 of 279. Its top-three rate reached 9.2% from 4.7% in September, and its rank-one rate reached 1.1% with 3 placements.

Clearcover's sustained upward trajectory is the cleanest signal in the category, with each month building on the previous reading without a single-month spike.

Highest-priority diagnostic: Whether Clearcover's gains are concentrated in Google AI Mode, where it is the top authority company in September, or increasingly spread across surfaces.

The General Series: Legacy Identity Holds at Zero

The General recorded 0.0% valid recommendation coverage for a second consecutive month in October, flat against September. Against the May 2026 baseline of 8.6%, the series registers a baseline-to-current decline recorded as significant, with a largest single-month move of 25.9 points between August and September.

The measured entity The General® registered 5.7% coverage in October, up 0.7 points from 5.0% in September. The General® now tracks a two-month upward streak with a current raw mention presence rate of 11.0% and a top-three rate of 3.2%.

The two tracked entities remain distinct. The General's flat 0.0% reading and The General®'s 5.7% reading should be read together as the brand's total October footprint rather than as the legacy identity's ongoing decline.

Highest-priority diagnostic: Whether The General brand's combined footprint across both identities has stabilised near the September-October range or remains below the 25.9% level registered in August 2026.

Kemper Auto: Fourth Consecutive Monthly Gain

Kemper Auto reached 2.8% valid recommendation coverage in October, up 0.3 points from 2.5% in September and 2.8 points from its 0.0% May 2026 baseline. This is Kemper's fourth consecutive monthly gain.

Presence held at 4.6% raw mention presence, with the brand appearing in 13 of 282 observations versus 14 of 279. Its top-three rate was 1.1%, and its rank-one rate was 0.0% for the second consecutive month. Valid recommendation count was 8 in October versus 7 in September.

Kemper's gains in coverage are real but its rank-one performance remains flat, indicating responsiveness is accruing at lower positions in recommendation shortlists.

Highest-priority diagnostic: Which prompts produce Kemper's top-three placements, and whether the brand's rank-one rate remains constrained by larger-budget competitors appearing first.

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters produced qualified observations in October?
• What can this benchmark not yet say about how AI systems weigh price or compare insurers head-to-head?

All 282 qualified observations in October fell into the Brand Recommendation cluster, consistent with September's 279 and earlier months. The comparison and pricing clusters registered no qualified observations in the current benchmark.

The public benchmark cannot yet answer price-sensitive or head-to-head comparison questions directly. The October evidence shows which brands AI systems recommend in general discovery and consideration prompts, but not how those systems weigh cost, coverage value, or comparative trade-offs across multiple options in a single answer. Brands that compete primarily on price or feature differentiation are therefore not yet visible in this benchmark's buyer-intent mix. The 13 inconsistency alerts included in this report do surface pricing and ownership claims from the underlying AI answers, but those claims are not part of the qualified recommendation benchmark set.

Brand Opportunity Summary

Questions This Section Answers

• What is the highest-priority diagnostic for each brand in the October benchmark?
• Which brands are rising, declining, or holding flat in AI recommendation coverage?

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 CiteWorks Studio's AI Visibility Industry Market 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

• Movement in this benchmark is directional, not causal. A brand's coverage change identifies where attention is warranted, not why the change occurred.
• The qualified denominator (282 observations in October) differs from the raw collection (800 prompts). Brand percentages are calculated within the qualified set only.
• The General identity split means the October comparison reflects two tracked entities where a single entity was tracked through August 2026. The General and The General® should be read together for the brand's total footprint.
• Small-count brands (Branch Insurance at 0.4%, Elephant Insurance at 1.4%, Kemper Auto at 2.8%) carry higher measurement uncertainty; their movements should be interpreted with care.
• The July 2026 instrument change expanded the collection universe and added a sixth surface family; baseline-to-current comparisons reflect this broader measurement context.
• October's 13 critical or high-severity inconsistency alerts are separate from the qualification pipeline. They surface contradictions in the underlying AI answers but are not part of the qualified recommendation benchmark set.

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 specific questions: which high-intent prompts does a brand win, which competitor takes the recommendation when a brand loses, what attributes do AI systems associate with each option, and which external sources shape those answers? The benchmark measures the outcome, not the mechanism, and October's 13 inconsistency alerts show that even basic facts about ownership and pricing remain contested across AI surfaces.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. That is the step from knowing where a brand stands to knowing what to do about it.

Request an AI visibility audit