{ "title": "Sonarvue AI Brand Accuracy Index 2026", "publishedDate": "2026-07-16", "status": "complete", "methodologyUrl": "https://sonarvue.com/blog/ai-brand-accuracy-index/", "sample": { "companies": 100, "domainsAssessed": 212, "sizeBands": { "1-10": 25, "101-200": 25, "11-50": 25, "51-100": 25 }, "countries": { "Canada": 13, "United States": 81, "United Kingdom": 4, "Australia": 2 }, "verifiedOfficialFacts": 1048 }, "collection": { "answerCaptures": 300, "claimsScored": 2862, "searchEnabledClaimsScored": 1999, "searchEnabledModels": [ "GPT 5.5", "Gemini 3 Flash" ], "knowledgeOnlyBaselines": [ "Claude Fable 5" ], "reportedSourceAttribution": [ { "modelId": "gpt-5-5", "model": "GPT 5.5", "assertedClaims": 775, "claimsWithReportedSource": 775, "reportedSourceRate": 100, "claimsWithOfficialSource": 750, "officialSourceRate": 96.8, "claimsWithThirdPartySource": 25, "thirdPartySourceRate": 3.2, "uniqueReportedSourceDomains": 126, "uniqueThirdPartySourceDomains": 8 }, { "modelId": "claude-fable-5", "model": "Claude Fable 5", "assertedClaims": 359, "claimsWithReportedSource": 260, "reportedSourceRate": 72.4, "claimsWithOfficialSource": 255, "officialSourceRate": 71, "claimsWithThirdPartySource": 5, "thirdPartySourceRate": 1.4, "uniqueReportedSourceDomains": 53, "uniqueThirdPartySourceDomains": 5 }, { "modelId": "gemini-3-flash", "model": "Gemini 3 Flash", "assertedClaims": 642, "claimsWithReportedSource": 641, "reportedSourceRate": 99.8, "claimsWithOfficialSource": 467, "officialSourceRate": 72.7, "claimsWithThirdPartySource": 174, "thirdPartySourceRate": 27.1, "uniqueReportedSourceDomains": 163, "uniqueThirdPartySourceDomains": 68 } ] }, "aggregate": { "verdicts": { "not_verifiable": 857, "supported": 552, "non_claim": 582, "contradicted": 8 }, "verifiedAccuracyPercent": 98.6, "contradictionRatePercent": 1.4, "evidenceCoveragePercent": 39.5, "judgeAgreement": { "companiesAudited": 20, "claimsCompared": 563, "matchingVerdicts": 528, "rate": 0.938, "perAnswerModel": [ { "modelId": "gpt-5-5", "model": "GPT 5.5", "claimsCompared": 202, "matchingVerdicts": 194, "rate": 0.96 }, { "modelId": "claude-fable-5", "model": "Claude Fable 5", "claimsCompared": 167, "matchingVerdicts": 157, "rate": 0.94 }, { "modelId": "gemini-3-flash", "model": "Gemini 3 Flash", "claimsCompared": 194, "matchingVerdicts": 177, "rate": 0.912 } ], "byPrimaryVerdict": [ { "verdict": "supported", "claimsCompared": 127, "matchingVerdicts": 126, "rate": 0.992 }, { "verdict": "not_verifiable", "claimsCompared": 192, "matchingVerdicts": 158, "rate": 0.823 }, { "verdict": "non_claim", "claimsCompared": 244, "matchingVerdicts": 244, "rate": 1 } ], "confusion": { "non_claim->non_claim": 244, "not_verifiable->not_verifiable": 158, "not_verifiable->supported": 34, "supported->not_verifiable": 1, "supported->supported": 126 } }, "perReportedSourceType": [ { "label": "official_domain", "claims": 1305, "supported": 514, "contradicted": 4, "notVerifiable": 699, "nonClaims": 88, "verifiedAccuracy": 99.2, "contradictionRate": 0.8, "evidenceCoverage": 42.6 }, { "label": "third_party", "claims": 206, "supported": 38, "contradicted": 4, "notVerifiable": 157, "nonClaims": 7, "verifiedAccuracy": 90.5, "contradictionRate": 9.5, "evidenceCoverage": 21.1 }, { "label": "no_reported_source", "claims": 488, "supported": 0, "contradicted": 0, "notVerifiable": 1, "nonClaims": 487, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": 0 } ], "perFieldAndReportedSourceType": [ { "label": "positioning:official_domain", "field": "positioning", "sourceType": "official_domain", "claims": 167, "supported": 82, "contradicted": 2, "notVerifiable": 83, "nonClaims": 0, "verifiedAccuracy": 97.6, "contradictionRate": 2.4, "evidenceCoverage": 50.3 }, { "label": "positioning:third_party", "field": "positioning", "sourceType": "third_party", "claims": 32, "supported": 10, "contradicted": 1, "notVerifiable": 21, "nonClaims": 0, "verifiedAccuracy": 90.9, "contradictionRate": 9.1, "evidenceCoverage": 34.4 }, { "label": "positioning:no_reported_source", "field": "positioning", "sourceType": "no_reported_source", "claims": 1, "supported": 0, "contradicted": 0, "notVerifiable": 0, "nonClaims": 1, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null }, { "label": "target_customer:official_domain", "field": "target_customer", "sourceType": "official_domain", "claims": 146, "supported": 76, "contradicted": 0, "notVerifiable": 70, "nonClaims": 0, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 52.1 }, { "label": "target_customer:third_party", "field": "target_customer", "sourceType": "third_party", "claims": 21, "supported": 1, "contradicted": 0, "notVerifiable": 20, "nonClaims": 0, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 4.8 }, { "label": "target_customer:no_reported_source", "field": "target_customer", "sourceType": "no_reported_source", "claims": 33, "supported": 0, "contradicted": 0, "notVerifiable": 1, "nonClaims": 32, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": 0 }, { "label": "capability:official_domain", "field": "capability", "sourceType": "official_domain", "claims": 515, "supported": 217, "contradicted": 0, "notVerifiable": 298, "nonClaims": 0, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 42.1 }, { "label": "capability:third_party", "field": "capability", "sourceType": "third_party", "claims": 73, "supported": 16, "contradicted": 0, "notVerifiable": 57, "nonClaims": 0, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 21.9 }, { "label": "capability:no_reported_source", "field": "capability", "sourceType": "no_reported_source", "claims": 2, "supported": 0, "contradicted": 0, "notVerifiable": 0, "nonClaims": 2, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null }, { "label": "integration:official_domain", "field": "integration", "sourceType": "official_domain", "claims": 314, "supported": 103, "contradicted": 0, "notVerifiable": 201, "nonClaims": 10, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 33.9 }, { "label": "integration:third_party", "field": "integration", "sourceType": "third_party", "claims": 45, "supported": 9, "contradicted": 0, "notVerifiable": 36, "nonClaims": 0, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 20 }, { "label": "integration:no_reported_source", "field": "integration", "sourceType": "no_reported_source", "claims": 50, "supported": 0, "contradicted": 0, "notVerifiable": 0, "nonClaims": 50, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null }, { "label": "starting_price:official_domain", "field": "starting_price", "sourceType": "official_domain", "claims": 61, "supported": 14, "contradicted": 1, "notVerifiable": 21, "nonClaims": 25, "verifiedAccuracy": 93.3, "contradictionRate": 6.7, "evidenceCoverage": 41.7 }, { "label": "starting_price:third_party", "field": "starting_price", "sourceType": "third_party", "claims": 16, "supported": 1, "contradicted": 2, "notVerifiable": 9, "nonClaims": 4, "verifiedAccuracy": 33.3, "contradictionRate": 66.7, "evidenceCoverage": 25 }, { "label": "starting_price:no_reported_source", "field": "starting_price", "sourceType": "no_reported_source", "claims": 123, "supported": 0, "contradicted": 0, "notVerifiable": 0, "nonClaims": 123, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null }, { "label": "free_plan:official_domain", "field": "free_plan", "sourceType": "official_domain", "claims": 44, "supported": 5, "contradicted": 1, "notVerifiable": 8, "nonClaims": 30, "verifiedAccuracy": 83.3, "contradictionRate": 16.7, "evidenceCoverage": 42.9 }, { "label": "free_plan:third_party", "field": "free_plan", "sourceType": "third_party", "claims": 12, "supported": 0, "contradicted": 0, "notVerifiable": 10, "nonClaims": 2, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": 0 }, { "label": "free_plan:no_reported_source", "field": "free_plan", "sourceType": "no_reported_source", "claims": 144, "supported": 0, "contradicted": 0, "notVerifiable": 0, "nonClaims": 144, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null }, { "label": "trial:official_domain", "field": "trial", "sourceType": "official_domain", "claims": 58, "supported": 17, "contradicted": 0, "notVerifiable": 18, "nonClaims": 23, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 48.6 }, { "label": "trial:third_party", "field": "trial", "sourceType": "third_party", "claims": 7, "supported": 1, "contradicted": 1, "notVerifiable": 4, "nonClaims": 1, "verifiedAccuracy": 50, "contradictionRate": 50, "evidenceCoverage": 33.3 }, { "label": "trial:no_reported_source", "field": "trial", "sourceType": "no_reported_source", "claims": 135, "supported": 0, "contradicted": 0, "notVerifiable": 0, "nonClaims": 135, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null } ] }, "historicalComparison": { "matchedCompanies": 100, "scoreContext": { "previousScoreVersion": 2, "currentScoreVersion": 3, "previousJudgeIds": [ "gpt-5-2" ], "currentJudgeIds": [ "gpt-5-5" ], "previousIndependentAudit": { "companiesAudited": 23, "auditModels": [ "claude-haiku-4-5" ], "claimsCompared": 603, "matchingVerdicts": 564, "rate": 0.935, "perAnswerModel": [ { "modelId": "gpt-5-2", "model": "GPT 5.2", "claimsCompared": 216, "matchingVerdicts": 205, "rate": 0.949 }, { "modelId": "claude-haiku-4-5", "model": "Claude Haiku 4.5", "claimsCompared": 163, "matchingVerdicts": 158, "rate": 0.969 }, { "modelId": "gemini-2.5-flash", "model": "Gemini 2.5 Flash", "claimsCompared": 224, "matchingVerdicts": 201, "rate": 0.897 } ] }, "currentIndependentAudit": { "companiesAudited": 20, "auditModels": [ "claude-fable-5" ], "claimsCompared": 563, "matchingVerdicts": 528, "rate": 0.938, "perAnswerModel": [ { "modelId": "gpt-5-5", "model": "GPT 5.5", "claimsCompared": 202, "matchingVerdicts": 194, "rate": 0.96 }, { "modelId": "claude-fable-5", "model": "Claude Fable 5", "claimsCompared": 167, "matchingVerdicts": 157, "rate": 0.94 }, { "modelId": "gemini-3-flash", "model": "Gemini 3 Flash", "claimsCompared": 194, "matchingVerdicts": 177, "rate": 0.912 } ] } }, "previousSearchEnabled": { "label": "GPT 5.2 + Gemini 2.5 Flash", "cohort": "previous_search_enabled", "companies": 100, "claims": 1924, "supported": 711, "contradicted": 10, "not_verifiable": 472, "non_claim": 731, "verifiedAccuracy": 98.6, "contradictionRate": 1.4, "evidenceCoverage": 60.4, "nonClaimRate": 38, "claimsPerCompany": 19.2, "substantiveClaimsPerCompany": 11.9 }, "currentSearchEnabled": { "label": "GPT 5.5 + Gemini 3 Flash", "cohort": "current_search_enabled", "companies": 100, "claims": 1999, "supported": 552, "contradicted": 8, "not_verifiable": 857, "non_claim": 582, "verifiedAccuracy": 98.6, "contradictionRate": 1.4, "evidenceCoverage": 39.5, "nonClaimRate": 29.1, "claimsPerCompany": 20, "substantiveClaimsPerCompany": 14.2 }, "delta": { "verifiedAccuracy": 0, "contradictionRate": 0, "evidenceCoverage": -20.9, "nonClaimRate": -8.9, "claimsPerCompany": 0.8, "substantiveClaimsPerCompany": 2.3 }, "captureBehavior": { "previousSearch": { "cohort": "previous_search_enabled", "companies": 100, "answers": 200, "claims": 1924, "assertedClaims": 1193, "unknownClaims": 731, "assertionRate": 62, "claimsPerAnswer": 9.6, "assertedClaimsPerAnswer": 6 }, "currentSearch": { "cohort": "current_search_enabled", "companies": 100, "answers": 200, "claims": 1999, "assertedClaims": 1417, "unknownClaims": 582, "assertionRate": 70.9, "claimsPerAnswer": 10, "assertedClaimsPerAnswer": 7.1 }, "delta": { "assertionRate": 8.9, "claimsPerAnswer": 0.4, "assertedClaimsPerAnswer": 1.1 }, "netAssertionIncrease": 224, "perField": [ { "field": "integration", "previous": { "cohort": "previous_search_enabled", "field": "integration", "answers": 200, "claims": 336, "assertedClaims": 232, "assertionRate": 69, "assertedClaimsPerAnswer": 1.2 }, "current": { "cohort": "current_search_enabled", "field": "integration", "answers": 200, "claims": 409, "assertedClaims": 349, "assertionRate": 85.3, "assertedClaimsPerAnswer": 1.7 }, "delta": { "assertedClaims": 117, "assertionRate": 16.3, "assertedClaimsPerAnswer": 0.5 }, "shareOfNetAssertionIncrease": 52.2 }, { "field": "target_customer", "previous": { "cohort": "previous_search_enabled", "field": "target_customer", "answers": 200, "claims": 200, "assertedClaims": 109, "assertionRate": 54.5, "assertedClaimsPerAnswer": 0.5 }, "current": { "cohort": "current_search_enabled", "field": "target_customer", "answers": 200, "claims": 200, "assertedClaims": 168, "assertionRate": 84, "assertedClaimsPerAnswer": 0.8 }, "delta": { "assertedClaims": 59, "assertionRate": 29.5, "assertedClaimsPerAnswer": 0.3 }, "shareOfNetAssertionIncrease": 26.3 }, { "field": "starting_price", "previous": { "cohort": "previous_search_enabled", "field": "starting_price", "answers": 200, "claims": 200, "assertedClaims": 27, "assertionRate": 13.5, "assertedClaimsPerAnswer": 0.1 }, "current": { "cohort": "current_search_enabled", "field": "starting_price", "answers": 200, "claims": 200, "assertedClaims": 48, "assertionRate": 24, "assertedClaimsPerAnswer": 0.2 }, "delta": { "assertedClaims": 21, "assertionRate": 10.5, "assertedClaimsPerAnswer": 0.1 }, "shareOfNetAssertionIncrease": 9.4 }, { "field": "free_plan", "previous": { "cohort": "previous_search_enabled", "field": "free_plan", "answers": 200, "claims": 200, "assertedClaims": 12, "assertionRate": 6, "assertedClaimsPerAnswer": 0.1 }, "current": { "cohort": "current_search_enabled", "field": "free_plan", "answers": 200, "claims": 200, "assertedClaims": 24, "assertionRate": 12, "assertedClaimsPerAnswer": 0.1 }, "delta": { "assertedClaims": 12, "assertionRate": 6, "assertedClaimsPerAnswer": 0 }, "shareOfNetAssertionIncrease": 5.4 }, { "field": "trial", "previous": { "cohort": "previous_search_enabled", "field": "trial", "answers": 200, "claims": 200, "assertedClaims": 30, "assertionRate": 15, "assertedClaimsPerAnswer": 0.1 }, "current": { "cohort": "current_search_enabled", "field": "trial", "answers": 200, "claims": 200, "assertedClaims": 41, "assertionRate": 20.5, "assertedClaimsPerAnswer": 0.2 }, "delta": { "assertedClaims": 11, "assertionRate": 5.5, "assertedClaimsPerAnswer": 0.1 }, "shareOfNetAssertionIncrease": 4.9 }, { "field": "positioning", "previous": { "cohort": "previous_search_enabled", "field": "positioning", "answers": 200, "claims": 200, "assertedClaims": 197, "assertionRate": 98.5, "assertedClaimsPerAnswer": 1 }, "current": { "cohort": "current_search_enabled", "field": "positioning", "answers": 200, "claims": 200, "assertedClaims": 199, "assertionRate": 99.5, "assertedClaimsPerAnswer": 1 }, "delta": { "assertedClaims": 2, "assertionRate": 1, "assertedClaimsPerAnswer": 0 }, "shareOfNetAssertionIncrease": 0.9 }, { "field": "capability", "previous": { "cohort": "previous_search_enabled", "field": "capability", "answers": 200, "claims": 588, "assertedClaims": 586, "assertionRate": 99.7, "assertedClaimsPerAnswer": 2.9 }, "current": { "cohort": "current_search_enabled", "field": "capability", "answers": 200, "claims": 590, "assertedClaims": 588, "assertionRate": 99.7, "assertedClaimsPerAnswer": 2.9 }, "delta": { "assertedClaims": 2, "assertionRate": 0, "assertedClaimsPerAnswer": 0 }, "shareOfNetAssertionIncrease": 0.9 } ], "perModel": { "gpt": { "previous": { "modelId": "gpt-5-2", "model": "GPT 5.2", "companies": 100, "answers": 100, "claims": 964, "assertedClaims": 590, "unknownClaims": 374, "assertionRate": 61.2, "claimsPerAnswer": 9.6, "assertedClaimsPerAnswer": 5.9 }, "current": { "modelId": "gpt-5-5", "model": "GPT 5.5", "companies": 100, "answers": 100, "claims": 1037, "assertedClaims": 775, "unknownClaims": 262, "assertionRate": 74.7, "claimsPerAnswer": 10.4, "assertedClaimsPerAnswer": 7.8 }, "delta": { "assertionRate": 13.5, "claimsPerAnswer": 0.8, "assertedClaimsPerAnswer": 1.9 }, "transitions": { "perField": { "positioning": { "claimToClaim": 99, "claimToUnknown": 0, "unknownToClaim": 1, "unknownToUnknown": 0 }, "capability": { "claimToClaim": 98, "claimToUnknown": 0, "unknownToClaim": 2, "unknownToUnknown": 0 }, "target_customer": { "claimToClaim": 45, "claimToUnknown": 1, "unknownToClaim": 50, "unknownToUnknown": 4 }, "starting_price": { "claimToClaim": 14, "claimToUnknown": 2, "unknownToClaim": 15, "unknownToUnknown": 69 }, "free_plan": { "claimToClaim": 2, "claimToUnknown": 1, "unknownToClaim": 11, "unknownToUnknown": 86 }, "trial": { "claimToClaim": 14, "claimToUnknown": 3, "unknownToClaim": 10, "unknownToUnknown": 73 }, "integration": { "claimToClaim": 46, "claimToUnknown": 3, "unknownToClaim": 31, "unknownToUnknown": 20 } }, "newlyAssertedCompanyFields": 120, "newlyAbstainedCompanyFields": 10 } }, "gemini": { "previous": { "modelId": "gemini-2.5-flash", "model": "Gemini 2.5 Flash", "companies": 100, "answers": 100, "claims": 960, "assertedClaims": 603, "unknownClaims": 357, "assertionRate": 62.8, "claimsPerAnswer": 9.6, "assertedClaimsPerAnswer": 6 }, "current": { "modelId": "gemini-3-flash", "model": "Gemini 3 Flash", "companies": 100, "answers": 100, "claims": 962, "assertedClaims": 642, "unknownClaims": 320, "assertionRate": 66.7, "claimsPerAnswer": 9.6, "assertedClaimsPerAnswer": 6.4 }, "delta": { "assertionRate": 3.9, "claimsPerAnswer": 0, "assertedClaimsPerAnswer": 0.4 }, "transitions": { "perField": { "positioning": { "claimToClaim": 97, "claimToUnknown": 1, "unknownToClaim": 2, "unknownToUnknown": 0 }, "capability": { "claimToClaim": 98, "claimToUnknown": 2, "unknownToClaim": 0, "unknownToUnknown": 0 }, "target_customer": { "claimToClaim": 48, "claimToUnknown": 15, "unknownToClaim": 25, "unknownToUnknown": 12 }, "starting_price": { "claimToClaim": 6, "claimToUnknown": 5, "unknownToClaim": 13, "unknownToUnknown": 76 }, "free_plan": { "claimToClaim": 4, "claimToUnknown": 5, "unknownToClaim": 7, "unknownToUnknown": 84 }, "trial": { "claimToClaim": 5, "claimToUnknown": 8, "unknownToClaim": 12, "unknownToUnknown": 75 }, "integration": { "claimToClaim": 41, "claimToUnknown": 8, "unknownToClaim": 22, "unknownToUnknown": 29 } }, "newlyAssertedCompanyFields": 81, "newlyAbstainedCompanyFields": 44 } }, "claude": { "previous": { "modelId": "claude-haiku-4-5", "model": "Claude Haiku 4.5", "companies": 100, "answers": 100, "claims": 711, "assertedClaims": 38, "unknownClaims": 673, "assertionRate": 5.3, "claimsPerAnswer": 7.1, "assertedClaimsPerAnswer": 0.4 }, "current": { "modelId": "claude-fable-5", "model": "Claude Fable 5", "companies": 100, "answers": 100, "claims": 863, "assertedClaims": 359, "unknownClaims": 504, "assertionRate": 41.6, "claimsPerAnswer": 8.6, "assertedClaimsPerAnswer": 3.6 }, "delta": { "assertionRate": 36.3, "claimsPerAnswer": 1.5, "assertedClaimsPerAnswer": 3.2 }, "transitions": { "perField": { "positioning": { "claimToClaim": 7, "claimToUnknown": 0, "unknownToClaim": 59, "unknownToUnknown": 34 }, "capability": { "claimToClaim": 5, "claimToUnknown": 0, "unknownToClaim": 58, "unknownToUnknown": 37 }, "target_customer": { "claimToClaim": 4, "claimToUnknown": 0, "unknownToClaim": 47, "unknownToUnknown": 49 }, "starting_price": { "claimToClaim": 0, "claimToUnknown": 2, "unknownToClaim": 0, "unknownToUnknown": 98 }, "free_plan": { "claimToClaim": 0, "claimToUnknown": 2, "unknownToClaim": 0, "unknownToUnknown": 98 }, "trial": { "claimToClaim": 0, "claimToUnknown": 2, "unknownToClaim": 1, "unknownToUnknown": 97 }, "integration": { "claimToClaim": 4, "claimToUnknown": 1, "unknownToClaim": 23, "unknownToUnknown": 72 } }, "newlyAssertedCompanyFields": 188, "newlyAbstainedCompanyFields": 7 } } }, "pairedCompanies": { "companies": 100, "companiesWithMoreAssertions": 71, "companiesWithFewerAssertions": 17, "companiesUnchanged": 12, "medianClaimDeltaPerCompany": 2, "rows": [ { "domain": "cyphercor.com", "previousAssertedClaims": 12, "currentAssertedClaims": 18, "delta": 6 }, { "domain": "nowsecure.com", "previousAssertedClaims": 9, "currentAssertedClaims": 16, "delta": 7 }, { "domain": "orangehook.com", "previousAssertedClaims": 6, "currentAssertedClaims": 10, "delta": 4 }, { "domain": "cm-labs.com", "previousAssertedClaims": 10, "currentAssertedClaims": 15, "delta": 5 }, { "domain": "ampletrails.com", "previousAssertedClaims": 15, "currentAssertedClaims": 15, "delta": 0 }, { "domain": "buyatab.com", "previousAssertedClaims": 11, "currentAssertedClaims": 12, "delta": 1 }, { "domain": "ihie.org", "previousAssertedClaims": 13, "currentAssertedClaims": 14, "delta": 1 }, { "domain": "revelation.com", "previousAssertedClaims": 11, "currentAssertedClaims": 13, "delta": 2 }, { "domain": "captiv8.io", "previousAssertedClaims": 18, "currentAssertedClaims": 18, "delta": 0 }, { "domain": "wealthhubsolutions.com", "previousAssertedClaims": 13, "currentAssertedClaims": 16, "delta": 3 }, { "domain": "bizfitech.com", "previousAssertedClaims": 10, "currentAssertedClaims": 12, "delta": 2 }, { "domain": "veritone.com", "previousAssertedClaims": 10, "currentAssertedClaims": 12, "delta": 2 }, { "domain": "cgcookie.com", "previousAssertedClaims": 11, "currentAssertedClaims": 17, "delta": 6 }, { "domain": "honeybook.com", "previousAssertedClaims": 21, "currentAssertedClaims": 21, "delta": 0 }, { "domain": "bebee.com", "previousAssertedClaims": 10, "currentAssertedClaims": 14, "delta": 4 }, { "domain": "armor.com", "previousAssertedClaims": 11, "currentAssertedClaims": 17, "delta": 6 }, { "domain": "whiteecho.com", "previousAssertedClaims": 8, "currentAssertedClaims": 13, "delta": 5 }, { "domain": "emergensys.net", "previousAssertedClaims": 10, "currentAssertedClaims": 15, "delta": 5 }, { "domain": "framebridge.com", "previousAssertedClaims": 8, "currentAssertedClaims": 11, "delta": 3 }, { "domain": "maxservicesgroup.com", "previousAssertedClaims": 8, "currentAssertedClaims": 7, "delta": -1 }, { "domain": "focusschoolsoftware.com", "previousAssertedClaims": 13, "currentAssertedClaims": 16, "delta": 3 }, { "domain": "protocol.ai", "previousAssertedClaims": 8, "currentAssertedClaims": 12, "delta": 4 }, { "domain": "pax.us", "previousAssertedClaims": 9, "currentAssertedClaims": 11, "delta": 2 }, { "domain": "apexturbine.com", "previousAssertedClaims": 15, "currentAssertedClaims": 13, "delta": -2 }, { "domain": "systemman.com", "previousAssertedClaims": 13, "currentAssertedClaims": 16, "delta": 3 }, { "domain": "educationgalaxy.com", "previousAssertedClaims": 13, "currentAssertedClaims": 15, "delta": 2 }, { "domain": "radiusnetworks.com", "previousAssertedClaims": 10, "currentAssertedClaims": 14, "delta": 4 }, { "domain": "riverside.com", "previousAssertedClaims": 10, "currentAssertedClaims": 15, "delta": 5 }, { "domain": "omimattress.com", "previousAssertedClaims": 11, "currentAssertedClaims": 10, "delta": -1 }, { "domain": "interim-cfos.com", "previousAssertedClaims": 10, "currentAssertedClaims": 9, "delta": -1 }, { "domain": "placespeak.com", "previousAssertedClaims": 12, "currentAssertedClaims": 17, "delta": 5 }, { "domain": "appliedts.net", "previousAssertedClaims": 5, "currentAssertedClaims": 10, "delta": 5 }, { "domain": "innovaccer.com", "previousAssertedClaims": 18, "currentAssertedClaims": 15, "delta": -3 }, { "domain": "quartzy.com", "previousAssertedClaims": 18, "currentAssertedClaims": 19, "delta": 1 }, { "domain": "shaleexperts.com", "previousAssertedClaims": 12, "currentAssertedClaims": 11, "delta": -1 }, { "domain": "simpplr.com", "previousAssertedClaims": 15, "currentAssertedClaims": 19, "delta": 4 }, { "domain": "towbook.com", "previousAssertedClaims": 16, "currentAssertedClaims": 19, "delta": 3 }, { "domain": "noverant.com", "previousAssertedClaims": 10, "currentAssertedClaims": 12, "delta": 2 }, { "domain": "atlas.co", "previousAssertedClaims": 12, "currentAssertedClaims": 16, "delta": 4 }, { "domain": "bna-inc.com", "previousAssertedClaims": 10, "currentAssertedClaims": 14, "delta": 4 }, { "domain": "later.com", "previousAssertedClaims": 17, "currentAssertedClaims": 19, "delta": 2 }, { "domain": "helmoperations.com", "previousAssertedClaims": 13, "currentAssertedClaims": 11, "delta": -2 }, { "domain": "solutionstream.com", "previousAssertedClaims": 8, "currentAssertedClaims": 11, "delta": 3 }, { "domain": "ordoro.com", "previousAssertedClaims": 17, "currentAssertedClaims": 20, "delta": 3 }, { "domain": "aha.io", "previousAssertedClaims": 18, "currentAssertedClaims": 19, "delta": 1 }, { "domain": "mindtickle.com", "previousAssertedClaims": 16, "currentAssertedClaims": 16, "delta": 0 }, { "domain": "caicorp.com", "previousAssertedClaims": 11, "currentAssertedClaims": 15, "delta": 4 }, { "domain": "queue-it.com", "previousAssertedClaims": 12, "currentAssertedClaims": 18, "delta": 6 }, { "domain": "onestreamsoftware.com", "previousAssertedClaims": 11, "currentAssertedClaims": 14, "delta": 3 }, { "domain": "merlincryption.com", "previousAssertedClaims": 10, "currentAssertedClaims": 11, "delta": 1 }, { "domain": "founderscard.com", "previousAssertedClaims": 13, "currentAssertedClaims": 11, "delta": -2 }, { "domain": "techsophy.com", "previousAssertedClaims": 4, "currentAssertedClaims": 14, "delta": 10 }, { "domain": "jsitelecom.com", "previousAssertedClaims": 10, "currentAssertedClaims": 13, "delta": 3 }, { "domain": "kiddom.co", "previousAssertedClaims": 13, "currentAssertedClaims": 16, "delta": 3 }, { "domain": "caspio.com", "previousAssertedClaims": 16, "currentAssertedClaims": 19, "delta": 3 }, { "domain": "avitpartners.com", "previousAssertedClaims": 10, "currentAssertedClaims": 11, "delta": 1 }, { "domain": "logisense.com", "previousAssertedClaims": 13, "currentAssertedClaims": 17, "delta": 4 }, { "domain": "t5datacenters.com", "previousAssertedClaims": 10, "currentAssertedClaims": 10, "delta": 0 }, { "domain": "upstreamworks.com", "previousAssertedClaims": 11, "currentAssertedClaims": 14, "delta": 3 }, { "domain": "tbaytel.net", "previousAssertedClaims": 11, "currentAssertedClaims": 15, "delta": 4 }, { "domain": "cpitelecom.net", "previousAssertedClaims": 15, "currentAssertedClaims": 16, "delta": 1 }, { "domain": "elevate.blue", "previousAssertedClaims": 6, "currentAssertedClaims": 10, "delta": 4 }, { "domain": "shared.com", "previousAssertedClaims": 10, "currentAssertedClaims": 9, "delta": -1 }, { "domain": "lakeba.com", "previousAssertedClaims": 7, "currentAssertedClaims": 11, "delta": 4 }, { "domain": "cliptraining.com", "previousAssertedClaims": 15, "currentAssertedClaims": 14, "delta": -1 }, { "domain": "tails.com", "previousAssertedClaims": 11, "currentAssertedClaims": 10, "delta": -1 }, { "domain": "lucidworks.com", "previousAssertedClaims": 11, "currentAssertedClaims": 19, "delta": 8 }, { "domain": "interviewing.io", "previousAssertedClaims": 10, "currentAssertedClaims": 11, "delta": 1 }, { "domain": "livepos.com", "previousAssertedClaims": 16, "currentAssertedClaims": 13, "delta": -3 }, { "domain": "newstore.com", "previousAssertedClaims": 15, "currentAssertedClaims": 11, "delta": -4 }, { "domain": "noredink.com", "previousAssertedClaims": 12, "currentAssertedClaims": 18, "delta": 6 }, { "domain": "holtinternational.org", "previousAssertedClaims": 9, "currentAssertedClaims": 11, "delta": 2 }, { "domain": "bombbomb.com", "previousAssertedClaims": 20, "currentAssertedClaims": 20, "delta": 0 }, { "domain": "goamplifi.com", "previousAssertedClaims": 12, "currentAssertedClaims": 14, "delta": 2 }, { "domain": "maxta.com", "previousAssertedClaims": 11, "currentAssertedClaims": 16, "delta": 5 }, { "domain": "factr.com", "previousAssertedClaims": 9, "currentAssertedClaims": 12, "delta": 3 }, { "domain": "emplay.net", "previousAssertedClaims": 9, "currentAssertedClaims": 16, "delta": 7 }, { "domain": "jamcracker.com", "previousAssertedClaims": 16, "currentAssertedClaims": 17, "delta": 1 }, { "domain": "buster.com", "previousAssertedClaims": 9, "currentAssertedClaims": 9, "delta": 0 }, { "domain": "layer.com", "previousAssertedClaims": 15, "currentAssertedClaims": 10, "delta": -5 }, { "domain": "4crisk.ai", "previousAssertedClaims": 13, "currentAssertedClaims": 13, "delta": 0 }, { "domain": "iopipe.com", "previousAssertedClaims": 11, "currentAssertedClaims": 18, "delta": 7 }, { "domain": "nevastar.com", "previousAssertedClaims": 10, "currentAssertedClaims": 9, "delta": -1 }, { "domain": "appointment-plus.com", "previousAssertedClaims": 13, "currentAssertedClaims": 16, "delta": 3 }, { "domain": "atheerair.com", "previousAssertedClaims": 11, "currentAssertedClaims": 16, "delta": 5 }, { "domain": "concepture.com", "previousAssertedClaims": 15, "currentAssertedClaims": 16, "delta": 1 }, { "domain": "ordergroove.com", "previousAssertedClaims": 12, "currentAssertedClaims": 16, "delta": 4 }, { "domain": "knotel.com", "previousAssertedClaims": 10, "currentAssertedClaims": 10, "delta": 0 }, { "domain": "inspectionsupport.net", "previousAssertedClaims": 14, "currentAssertedClaims": 19, "delta": 5 }, { "domain": "scratchpadstudios.com", "previousAssertedClaims": 10, "currentAssertedClaims": 5, "delta": -5 }, { "domain": "repugen.com", "previousAssertedClaims": 17, "currentAssertedClaims": 16, "delta": -1 }, { "domain": "catchafire.org", "previousAssertedClaims": 11, "currentAssertedClaims": 11, "delta": 0 }, { "domain": "aptible.com", "previousAssertedClaims": 15, "currentAssertedClaims": 20, "delta": 5 }, { "domain": "as-software.com", "previousAssertedClaims": 14, "currentAssertedClaims": 14, "delta": 0 }, { "domain": "onspring.com", "previousAssertedClaims": 16, "currentAssertedClaims": 16, "delta": 0 }, { "domain": "biom.io", "previousAssertedClaims": 14, "currentAssertedClaims": 16, "delta": 2 }, { "domain": "cynergy.pro", "previousAssertedClaims": 10, "currentAssertedClaims": 16, "delta": 6 }, { "domain": "go-2-learn.com", "previousAssertedClaims": 8, "currentAssertedClaims": 9, "delta": 1 }, { "domain": "aisleplanner.com", "previousAssertedClaims": 12, "currentAssertedClaims": 19, "delta": 7 }, { "domain": "southernfibernet.com", "previousAssertedClaims": 10, "currentAssertedClaims": 12, "delta": 2 } ] } }, "sourceBehavior": { "previousSearch": { "cohort": "previous_search_enabled", "assertedClaims": 1196, "claimsWithReportedSource": 1196, "reportedSourceRate": 100, "claimsWithClassifiableSource": 906, "classifiableSourceRate": 75.8, "claimsWithOpaqueRetrievalRedirect": 290, "opaqueRetrievalRedirectRate": 24.2, "claimsWithOfficialSource": 840, "officialSourceRate": 92.7, "claimsWithThirdPartySource": 66, "thirdPartySourceRate": 7.3, "uniqueThirdPartySourceDomains": 14 }, "currentSearch": { "cohort": "current_search_enabled", "assertedClaims": 1417, "claimsWithReportedSource": 1416, "reportedSourceRate": 99.9, "claimsWithClassifiableSource": 1416, "classifiableSourceRate": 99.9, "claimsWithOpaqueRetrievalRedirect": 0, "opaqueRetrievalRedirectRate": 0, "claimsWithOfficialSource": 1217, "officialSourceRate": 85.9, "claimsWithThirdPartySource": 199, "thirdPartySourceRate": 14.1, "uniqueThirdPartySourceDomains": 74 }, "ownershipComparable": false, "delta": { "reportedSourceRate": -0.1, "classifiableSourceRate": 24.1, "opaqueRetrievalRedirectRate": -24.2, "officialSourceRate": null, "thirdPartySourceRate": null }, "perModel": { "gpt": { "previous": { "modelId": "gpt-5-2", "model": "GPT 5.2", "assertedClaims": 590, "claimsWithReportedSource": 590, "reportedSourceRate": 100, "claimsWithClassifiableSource": 590, "classifiableSourceRate": 100, "claimsWithOpaqueRetrievalRedirect": 0, "opaqueRetrievalRedirectRate": 0, "claimsWithOfficialSource": 546, "officialSourceRate": 92.5, "claimsWithThirdPartySource": 44, "thirdPartySourceRate": 7.5, "uniqueThirdPartySourceDomains": 4 }, "current": { "modelId": "gpt-5-5", "model": "GPT 5.5", "assertedClaims": 775, "claimsWithReportedSource": 775, "reportedSourceRate": 100, "claimsWithClassifiableSource": 775, "classifiableSourceRate": 100, "claimsWithOpaqueRetrievalRedirect": 0, "opaqueRetrievalRedirectRate": 0, "claimsWithOfficialSource": 750, "officialSourceRate": 96.8, "claimsWithThirdPartySource": 25, "thirdPartySourceRate": 3.2, "uniqueThirdPartySourceDomains": 8 }, "ownershipComparable": true, "delta": { "reportedSourceRate": 0, "classifiableSourceRate": 0, "opaqueRetrievalRedirectRate": 0, "officialSourceRate": 4.3, "thirdPartySourceRate": -4.3 } }, "gemini": { "previous": { "modelId": "gemini-2.5-flash", "model": "Gemini 2.5 Flash", "assertedClaims": 606, "claimsWithReportedSource": 606, "reportedSourceRate": 100, "claimsWithClassifiableSource": 316, "classifiableSourceRate": 52.1, "claimsWithOpaqueRetrievalRedirect": 290, "opaqueRetrievalRedirectRate": 47.9, "claimsWithOfficialSource": 294, "officialSourceRate": 93, "claimsWithThirdPartySource": 22, "thirdPartySourceRate": 7, "uniqueThirdPartySourceDomains": 10 }, "current": { "modelId": "gemini-3-flash", "model": "Gemini 3 Flash", "assertedClaims": 642, "claimsWithReportedSource": 641, "reportedSourceRate": 99.8, "claimsWithClassifiableSource": 641, "classifiableSourceRate": 99.8, "claimsWithOpaqueRetrievalRedirect": 0, "opaqueRetrievalRedirectRate": 0, "claimsWithOfficialSource": 467, "officialSourceRate": 72.9, "claimsWithThirdPartySource": 174, "thirdPartySourceRate": 27.1, "uniqueThirdPartySourceDomains": 68 }, "ownershipComparable": false, "delta": { "reportedSourceRate": -0.2, "classifiableSourceRate": 47.7, "opaqueRetrievalRedirectRate": -47.9, "officialSourceRate": null, "thirdPartySourceRate": null } }, "claude": { "previous": { "modelId": "claude-haiku-4-5", "model": "Claude Haiku 4.5", "assertedClaims": 38, "claimsWithReportedSource": 21, "reportedSourceRate": 55.3, "claimsWithClassifiableSource": 21, "classifiableSourceRate": 55.3, "claimsWithOpaqueRetrievalRedirect": 0, "opaqueRetrievalRedirectRate": 0, "claimsWithOfficialSource": 20, "officialSourceRate": 95.2, "claimsWithThirdPartySource": 1, "thirdPartySourceRate": 4.8, "uniqueThirdPartySourceDomains": 1 }, "current": { "modelId": "claude-fable-5", "model": "Claude Fable 5", "assertedClaims": 359, "claimsWithReportedSource": 260, "reportedSourceRate": 72.4, "claimsWithClassifiableSource": 260, "classifiableSourceRate": 72.4, "claimsWithOpaqueRetrievalRedirect": 0, "opaqueRetrievalRedirectRate": 0, "claimsWithOfficialSource": 255, "officialSourceRate": 98.1, "claimsWithThirdPartySource": 5, "thirdPartySourceRate": 1.9, "uniqueThirdPartySourceDomains": 5 }, "ownershipComparable": true, "delta": { "reportedSourceRate": 17.1, "classifiableSourceRate": 17.1, "opaqueRetrievalRedirectRate": 0, "officialSourceRate": 2.9, "thirdPartySourceRate": -2.9 } } } }, "errorPersistence": { "persistentDomains": [ "later.com", "shared.com" ], "clearedDomains": [ "cliptraining.com", "elevate.blue", "honeybook.com", "layer.com", "orangehook.com" ], "newErrorDomains": [ "helmoperations.com", "ordoro.com", "revelation.com", "towbook.com" ] }, "controlledSameJudgeRescore": { "generatedAt": "2026-07-15T21:32:23.136Z", "status": "controlled_comparison_complete", "judgeId": "claude-fable-5", "currentCompanies": 100, "companiesReady": 100, "companiesRescored": 100, "scoreRows": 3923, "errors": [], "automaticContradictions": 20, "adjudicatedAutomaticContradictions": 20, "unreviewedAutomaticContradictions": 0, "cohorts": { "previous": { "claims": 1924, "supported": 764, "contradicted": 10, "not_verifiable": 419, "non_claim": 731, "verifiedAccuracy": 98.7, "contradictionRate": 1.3, "evidenceCoverage": 64.9 }, "current": { "claims": 1999, "supported": 783, "contradicted": 8, "not_verifiable": 626, "non_claim": 582, "verifiedAccuracy": 99, "contradictionRate": 1, "evidenceCoverage": 55.8 } }, "perModel": [ { "cohort": "current", "modelId": "gpt-5-5", "model": "GPT 5.5", "claims": 1037, "supported": 451, "contradicted": 0, "not_verifiable": 324, "non_claim": 262, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 58.2 }, { "cohort": "previous", "modelId": "gemini-2.5-flash", "model": "Gemini 2.5 Flash", "claims": 960, "supported": 352, "contradicted": 5, "not_verifiable": 246, "non_claim": 357, "verifiedAccuracy": 98.6, "contradictionRate": 1.4, "evidenceCoverage": 59.2 }, { "cohort": "previous", "modelId": "gpt-5-2", "model": "GPT 5.2", "claims": 964, "supported": 412, "contradicted": 5, "not_verifiable": 173, "non_claim": 374, "verifiedAccuracy": 98.8, "contradictionRate": 1.2, "evidenceCoverage": 70.7 }, { "cohort": "current", "modelId": "gemini-3-flash", "model": "Gemini 3 Flash", "claims": 962, "supported": 332, "contradicted": 8, "not_verifiable": 302, "non_claim": 320, "verifiedAccuracy": 97.6, "contradictionRate": 2.4, "evidenceCoverage": 53 } ], "perField": [ { "field": "positioning", "previous": { "claims": 200, "supported": 167, "contradicted": 7, "not_verifiable": 23, "non_claim": 3, "verifiedAccuracy": 96, "contradictionRate": 4, "evidenceCoverage": 88.3 }, "current": { "claims": 200, "supported": 164, "contradicted": 3, "not_verifiable": 32, "non_claim": 1, "verifiedAccuracy": 98.2, "contradictionRate": 1.8, "evidenceCoverage": 83.9 } }, { "field": "target_customer", "previous": { "claims": 200, "supported": 76, "contradicted": 0, "not_verifiable": 33, "non_claim": 91, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 69.7 }, "current": { "claims": 200, "supported": 107, "contradicted": 0, "not_verifiable": 61, "non_claim": 32, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 63.7 } }, { "field": "capability", "previous": { "claims": 588, "supported": 395, "contradicted": 0, "not_verifiable": 191, "non_claim": 2, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 67.4 }, "current": { "claims": 590, "supported": 330, "contradicted": 0, "not_verifiable": 258, "non_claim": 2, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 56.1 } }, { "field": "integration", "previous": { "claims": 336, "supported": 102, "contradicted": 0, "not_verifiable": 130, "non_claim": 104, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 44 }, "current": { "claims": 409, "supported": 134, "contradicted": 0, "not_verifiable": 215, "non_claim": 60, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 38.4 } }, { "field": "starting_price", "previous": { "claims": 200, "supported": 11, "contradicted": 2, "not_verifiable": 14, "non_claim": 173, "verifiedAccuracy": 84.6, "contradictionRate": 15.4, "evidenceCoverage": 48.1 }, "current": { "claims": 200, "supported": 18, "contradicted": 3, "not_verifiable": 27, "non_claim": 152, "verifiedAccuracy": 85.7, "contradictionRate": 14.3, "evidenceCoverage": 43.8 } }, { "field": "free_plan", "previous": { "claims": 200, "supported": 2, "contradicted": 0, "not_verifiable": 10, "non_claim": 188, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 16.7 }, "current": { "claims": 200, "supported": 7, "contradicted": 1, "not_verifiable": 16, "non_claim": 176, "verifiedAccuracy": 87.5, "contradictionRate": 12.5, "evidenceCoverage": 33.3 } }, { "field": "trial", "previous": { "claims": 200, "supported": 11, "contradicted": 1, "not_verifiable": 18, "non_claim": 170, "verifiedAccuracy": 91.7, "contradictionRate": 8.3, "evidenceCoverage": 40 }, "current": { "claims": 200, "supported": 23, "contradicted": 1, "not_verifiable": 17, "non_claim": 159, "verifiedAccuracy": 95.8, "contradictionRate": 4.2, "evidenceCoverage": 58.5 } } ], "perReportedSourceType": [ { "sourceType": "official_domain", "previous": { "claims": 890, "supported": 542, "contradicted": 5, "not_verifiable": 273, "non_claim": 70, "verifiedAccuracy": 99.1, "contradictionRate": 0.9, "evidenceCoverage": 66.7 }, "current": { "claims": 1253, "supported": 678, "contradicted": 4, "not_verifiable": 485, "non_claim": 86, "verifiedAccuracy": 99.4, "contradictionRate": 0.6, "evidenceCoverage": 58.4 } }, { "sourceType": "third_party", "previous": { "claims": 399, "supported": 222, "contradicted": 5, "not_verifiable": 146, "non_claim": 26, "verifiedAccuracy": 97.8, "contradictionRate": 2.2, "evidenceCoverage": 60.9 }, "current": { "claims": 258, "supported": 104, "contradicted": 4, "not_verifiable": 141, "non_claim": 9, "verifiedAccuracy": 96.3, "contradictionRate": 3.7, "evidenceCoverage": 43.4 } }, { "sourceType": "no_reported_source", "previous": { "claims": 635, "supported": 0, "contradicted": 0, "not_verifiable": 0, "non_claim": 635, "verifiedAccuracy": null, "contradictionRate": null, "evidenceCoverage": null }, "current": { "claims": 488, "supported": 1, "contradicted": 0, "not_verifiable": 0, "non_claim": 487, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 100 } } ], "caveat": "Both cohorts use one blinded Claude Fable 5 judge and identical saved facts. Model answers were captured at different times and may reflect changing web content." }, "caveat": "The primary-score deltas use different score versions and judges. The controlled rescore uses one blinded Claude Fable 5 judge and identical saved facts for both cohorts. Archived opaque Vertex grounding redirects are excluded from source-ownership rates." }, "perModel": [ { "label": "GPT 5.5", "modelId": "gpt-5-5", "retrieval": "web_search", "comparable": true, "claims": 1037, "supported": 364, "contradicted": 0, "notVerifiable": 411, "nonClaims": 262, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 47 }, { "label": "Claude Fable 5", "modelId": "claude-fable-5", "retrieval": "none", "comparable": false, "claims": 863, "supported": 88, "contradicted": 0, "notVerifiable": 271, "nonClaims": 504, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 24.5 }, { "label": "Gemini 3 Flash", "modelId": "gemini-3-flash", "retrieval": "google_search", "comparable": true, "claims": 962, "supported": 188, "contradicted": 8, "notVerifiable": 446, "nonClaims": 320, "verifiedAccuracy": 95.9, "contradictionRate": 4.1, "evidenceCoverage": 30.5 } ], "perField": [ { "label": "positioning", "claims": 200, "supported": 92, "contradicted": 3, "notVerifiable": 104, "nonClaims": 1, "verifiedAccuracy": 96.8, "contradictionRate": 3.2, "evidenceCoverage": 47.7 }, { "label": "target_customer", "claims": 200, "supported": 77, "contradicted": 0, "notVerifiable": 91, "nonClaims": 32, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 45.8 }, { "label": "capability", "claims": 590, "supported": 233, "contradicted": 0, "notVerifiable": 355, "nonClaims": 2, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 39.6 }, { "label": "integration", "claims": 409, "supported": 112, "contradicted": 0, "notVerifiable": 237, "nonClaims": 60, "verifiedAccuracy": 100, "contradictionRate": 0, "evidenceCoverage": 32.1 }, { "label": "starting_price", "claims": 200, "supported": 15, "contradicted": 3, "notVerifiable": 30, "nonClaims": 152, "verifiedAccuracy": 83.3, "contradictionRate": 16.7, "evidenceCoverage": 37.5 }, { "label": "free_plan", "claims": 200, "supported": 5, "contradicted": 1, "notVerifiable": 18, "nonClaims": 176, "verifiedAccuracy": 83.3, "contradictionRate": 16.7, "evidenceCoverage": 25 }, { "label": "trial", "claims": 200, "supported": 18, "contradicted": 1, "notVerifiable": 22, "nonClaims": 159, "verifiedAccuracy": 94.7, "contradictionRate": 5.3, "evidenceCoverage": 46.3 } ], "manualReview": { "automatedContradictionsFlagged": 11, "automatedContradictionsAdjudicated": 11, "finalContradictionClaims": 8 }, "confirmedContradictions": [ { "company": "revelation software", "domain": "revelation.com", "model": "Gemini 3 Flash", "field": "positioning", "claim": "An online qualitative research platform for market insights.", "officialEvidence": [ { "field": "positioning", "value": "NoSQL MultiValue database development suite", "evidenceQuote": "Our flagship product OpenInsight is a NoSQL MultiValue database development suite", "evidenceUrl": "https://revelation.com/" } ], "adjudication": "The answer identifies Revelation Software as an online qualitative-research platform. The captured official homepage explicitly positions its flagship OpenInsight product as a NoSQL MultiValue database development suite." }, { "company": "towbook", "domain": "towbook.com", "model": "Gemini 3 Flash", "field": "starting_price", "claim": "$49 per month.", "officialEvidence": [ { "field": "starting_price", "value": "USD 109 per month", "evidenceQuote": "Basic $ 109 /mo", "evidenceUrl": "https://towbook.com/pricing" } ], "adjudication": "The answer states a $49 monthly starting price. The captured official pricing page explicitly lists the Basic plan at $109 per month." }, { "company": "towbook", "domain": "towbook.com", "model": "Gemini 3 Flash", "field": "trial", "claim": "Available 90-day free trial.", "officialEvidence": [ { "field": "trial", "value": "30-day free trial", "evidenceQuote": "Sign up for a 30-day free trial", "evidenceUrl": "https://towbook.com/" } ], "adjudication": "The answer states a 90-day free trial. The captured official homepage explicitly states a 30-day free trial." }, { "company": "later", "domain": "later.com", "model": "Gemini 3 Flash", "field": "starting_price", "claim": "Paid plans starting at $25 per month.", "officialEvidence": [ { "field": "starting_price", "value": "USD 18.75/month billed yearly", "evidenceQuote": "Starter For casual scheduling $ 18.75 USD/month* Billed yearly", "evidenceUrl": "https://later.com/pricing/" } ], "adjudication": "The answer presents $25 per month as the unqualified starting price. The captured official pricing page explicitly states that the Starter plan begins at $18.75 per month when billed yearly." }, { "company": "helm operations", "domain": "helmoperations.com", "model": "Gemini 3 Flash", "field": "positioning", "claim": "A platform for managing and automating containerized application deployments on Kubernetes.", "officialEvidence": [ { "field": "target_customer", "value": "Commercial vessel operators", "evidenceQuote": "building software that meets the real-world demands of commercial vessel operations", "evidenceUrl": "https://www.helmoperations.com/news/helm-operations-receives-us-coast-guard-recognition-for-digital-logbook-solution" }, { "field": "capability", "value": "Manage recurring maintenance routines", "evidenceQuote": "Helm CONNECT Maintenance allows you to manage your fleet’s recurring maintenance routines with ease.", "evidenceUrl": "https://www.helmoperations.com/maintenance-features/effortless-planned-maintenance" }, { "field": "capability", "value": "Digital logbook replacing paper-based record keeping", "evidenceQuote": "Helm CONNECT Logbook replaces paper-based record keeping with a digital solution", "evidenceUrl": "https://www.helmoperations.com/news/helm-operations-receives-us-coast-guard-recognition-for-digital-logbook-solution" } ], "adjudication": "The answer identifies Helm Operations as a Kubernetes deployment platform. The retained official facts position Helm CONNECT as maritime fleet-management software for vessel operations, fleet maintenance, and digital logbooks, which is a direct product-identity conflict." }, { "company": "ordoro inc", "domain": "ordoro.com", "model": "Gemini 3 Flash", "field": "starting_price", "claim": "$299/mo.", "officialEvidence": [ { "field": "starting_price", "value": "Shipping starts at $0/month", "evidenceQuote": "Shipping starts at $0/month (yep, free)", "evidenceUrl": "https://www.ordoro.com/" } ], "adjudication": "The answer presents $299 per month as Ordoro's unqualified starting price. The retained official pricing page shows Shipping starts at $0 per month, while $299 per month applies specifically to Dropshipping, so it is not Ordoro's entry price." }, { "company": "ordoro inc", "domain": "ordoro.com", "model": "Gemini 3 Flash", "field": "free_plan", "claim": "No free plan is offered.", "officialEvidence": [ { "field": "starting_price", "value": "Shipping starts at $0/month", "evidenceQuote": "Shipping starts at $0/month (yep, free)", "evidenceUrl": "https://www.ordoro.com/" } ], "adjudication": "The answer says no free plan is offered. The captured official pricing page explicitly advertises the Shipping product with 'Get started for FREE.'" }, { "company": "shared", "domain": "shared.com", "model": "Gemini 3 Flash", "field": "positioning", "claim": "Online content creator and digital publisher.", "officialEvidence": [ { "field": "positioning", "value": "White-label operating system for online-ordering businesses", "evidenceQuote": "Shared is the white-label operating system for online-ordering businesses", "evidenceUrl": "https://shared.com/" } ], "adjudication": "The answer identifies Shared as an online content creator and publisher. The captured official homepage explicitly positions Shared as a white-label operating system for online-ordering businesses." } ], "limitations": [ "The sample contains B2B software companies with 1 to 200 employees in the United States, Canada, United Kingdom, and Australia.", "Each model-company answer was captured once on 2026-07-15, so reruns may differ.", "Aggregate accuracy and evidence coverage include only search-enabled GPT 5.5 and Gemini 3 Flash captures.", "A claim absent from captured official evidence is not counted as wrong; it is marked not verifiable.", "Historical primary-score deltas use different score versions and judges; generation claims rely on the separate blinded same-judge rescore." ] }