{
  "dataset": "dixon.ai — the AI Reliability Scoreboard (documented errors and catches per model)",
  "description": "A running league table of how many specific, observable, falsifiable AI failures — and genuine catches — Ben Dixon has documented per model while using them on real investing decisions. A model appears once it has at least one documented entry; an assistant with no recorded failure or catch yet has no row. Every count is derived at build time from the findings and catches recorded on published posts; each model links back to the individual /lessons and /catches entries behind its tally.",
  "methodology": "This is a record of what happened in real tests, not a benchmark. The same prompt was not run an equal number of times against every model, the cases were not sampled at random, and the totals reflect what got written up — so they measure documented evidence, not relative reliability. Raw counts only; no rates or percentages are published because the sample is far too small to support them.",
  "disclaimer": "Small sample (low tens of cases). Treat these as documented cases, not a failure rate. A model with more documented errors here was not necessarily tested more, nor is it necessarily less reliable — it may simply have been used more, or failed in more screenshot-worthy ways. Human-verified by the author before publication.",
  "cite": "Quote freely with attribution to dixon.ai and a link to an entry URL. Author: Ben Dixon. Licence for this dataset: CC BY 4.0.",
  "site": "https://dixon.ai",
  "human_readable": "https://dixon.ai/scoreboard/",
  "sources": {
    "errors": "https://dixon.ai/evidence/wrong.json",
    "catches": "https://dixon.ai/evidence/caught.json"
  },
  "generated": "2026-09-08T17:57:26.956Z",
  "totals": {
    "models_scored": 7,
    "documented_errors": 55,
    "documented_catches": 50
  },
  "models": [
    {
      "tool": "perplexity",
      "label": "Perplexity",
      "documented_errors": 16,
      "documented_catches": 5,
      "errors_by_type": {
        "confabulated-own-transcript": 1,
        "false-confession-to-fabrication": 1,
        "stale-figure-as-current": 1,
        "misattributed-source": 1,
        "outdated-rule-stated-as-current": 1,
        "private-mode-identity-leak": 1,
        "partial-fabrication": 1,
        "internal-contradiction-wrong-lead": 1,
        "ignored-constraint": 1,
        "unit-error": 2,
        "inconsistent-5yr-returns": 1,
        "requested-measure-substitution": 1,
        "hedged-misattribution": 1,
        "dead-source-url": 1,
        "low-authority-source-led": 1
      },
      "errors_by_hallucination_type": {
        "faithfulness": 4,
        "factuality": 2
      },
      "errors_by_consequence": {
        "cosmetic-but-revealing": 5,
        "decision-affecting": 10
      },
      "catches_by_type": {
        "refused-false-confession": 1,
        "correct-source-attribution": 1,
        "honest-substitution": 1,
        "refused-to-quote-a-live-options-chain": 1,
        "refused-to-invent-a-citation": 1
      },
      "catches_by_impact": {
        "validation-only": 2,
        "decision-relevant": 3
      },
      "error_entries": [
        {
          "id": "perplexity-hedged-misattribution-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-hedged-misattribution-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "perplexity-requested-measure-substitution-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-requested-measure-substitution-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "perplexity-dead-source-url-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-dead-source-url-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "perplexity-confabulated-own-transcript-2026-07-22",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-confabulated-own-transcript-2026-07-22",
          "date": "2026-07-22"
        },
        {
          "id": "perplexity-false-confession-to-fabrication-2026-07-22",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-false-confession-to-fabrication-2026-07-22",
          "date": "2026-07-22"
        },
        {
          "id": "perplexity-partial-fabrication-2026-07-12",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-partial-fabrication-2026-07-12",
          "date": "2026-07-12"
        },
        {
          "id": "perplexity-stale-figure-as-current-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-stale-figure-as-current-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "perplexity-misattributed-source-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-misattributed-source-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "perplexity-private-mode-identity-leak-2026-07-06",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-private-mode-identity-leak-2026-07-06",
          "date": "2026-07-06"
        },
        {
          "id": "perplexity-internal-contradiction-wrong-lead-2026-06-26",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-internal-contradiction-wrong-lead-2026-06-26",
          "date": "2026-06-26"
        },
        {
          "id": "perplexity-low-authority-source-led-2026-06-20",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-low-authority-source-led-2026-06-20",
          "date": "2026-06-20"
        },
        {
          "id": "perplexity-outdated-rule-stated-as-current-2026-06-19",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-outdated-rule-stated-as-current-2026-06-19",
          "date": "2026-06-19"
        },
        {
          "id": "perplexity-inconsistent-5yr-returns-2026-06-13",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-inconsistent-5yr-returns-2026-06-13",
          "date": "2026-06-13"
        },
        {
          "id": "perplexity-ignored-constraint-2026-05-15",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-ignored-constraint-2026-05-15",
          "date": "2026-05-15"
        },
        {
          "id": "perplexity-unit-error-2026-05-14",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-unit-error-2026-05-14",
          "date": "2026-05-14"
        },
        {
          "id": "perplexity-unit-error-2026-05-14",
          "url": "https://dixon.ai/evidence/?outcome=wrong#perplexity-unit-error-2026-05-14",
          "date": "2026-05-14"
        }
      ],
      "catch_entries": [
        {
          "id": "perplexity-refused-to-quote-a-live-options-chain-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=caught#perplexity-refused-to-quote-a-live-options-chain-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "perplexity-refused-to-invent-a-citation-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=caught#perplexity-refused-to-invent-a-citation-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "perplexity-refused-false-confession-2026-07-22",
          "url": "https://dixon.ai/evidence/?outcome=caught#perplexity-refused-false-confession-2026-07-22",
          "date": "2026-07-22"
        },
        {
          "id": "perplexity-correct-source-attribution-2026-06-20",
          "url": "https://dixon.ai/evidence/?outcome=caught#perplexity-correct-source-attribution-2026-06-20",
          "date": "2026-06-20"
        },
        {
          "id": "perplexity-honest-substitution-2026-06-18",
          "url": "https://dixon.ai/evidence/?outcome=caught#perplexity-honest-substitution-2026-06-18",
          "date": "2026-06-18"
        }
      ]
    },
    {
      "tool": "gemini",
      "label": "Gemini",
      "documented_errors": 13,
      "documented_catches": 7,
      "errors_by_type": {
        "opaque-provenance": 1,
        "incomplete-jurisdiction": 1,
        "unsupported-premium-estimates": 1,
        "fabricated-interface-element": 1,
        "vague-source-attribution": 1,
        "fabrication": 1,
        "partial-fabrication": 1,
        "wrong-entity-audit": 1,
        "stale-memory-as-current": 1,
        "misattribution": 1,
        "unprompted-cross-conversation-memory": 1,
        "chat-prose-spliced-into-code": 1,
        "unguarded-storage-call-kills-the-page": 1
      },
      "errors_by_hallucination_type": {
        "factuality": 1,
        "faithfulness": 3
      },
      "errors_by_consequence": {
        "decision-affecting": 9,
        "cosmetic-but-revealing": 3
      },
      "catches_by_type": {
        "refused-false-confession": 1,
        "exact-closing-price-correctly-dated": 1,
        "false-premise-caught": 1,
        "held-correct-answer-under-pushback": 1,
        "entity-overlap-risk": 1,
        "stale-data-flag": 1,
        "premise-challenge": 1
      },
      "catches_by_impact": {
        "validation-only": 1,
        "insight-relevant": 3,
        "decision-relevant": 3
      },
      "error_entries": [
        {
          "id": "gemini-chat-prose-spliced-into-code-2026-08-06",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-chat-prose-spliced-into-code-2026-08-06",
          "date": "2026-08-06"
        },
        {
          "id": "gemini-unguarded-storage-call-kills-the-page-2026-08-06",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-unguarded-storage-call-kills-the-page-2026-08-06",
          "date": "2026-08-06"
        },
        {
          "id": "gemini-opaque-provenance-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-opaque-provenance-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "gemini-incomplete-jurisdiction-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-incomplete-jurisdiction-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "gemini-misattribution-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-misattribution-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "gemini-unprompted-cross-conversation-memory-2026-06-13",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-unprompted-cross-conversation-memory-2026-06-13",
          "date": "2026-06-13"
        },
        {
          "id": "gemini-fabricated-interface-element-2026-06-12",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-fabricated-interface-element-2026-06-12",
          "date": "2026-06-12"
        },
        {
          "id": "gemini-vague-source-attribution-2026-06-12",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-vague-source-attribution-2026-06-12",
          "date": "2026-06-12"
        },
        {
          "id": "gemini-wrong-entity-audit-2026-06-10",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-wrong-entity-audit-2026-06-10",
          "date": "2026-06-10"
        },
        {
          "id": "gemini-stale-memory-as-current-2026-06-10",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-stale-memory-as-current-2026-06-10",
          "date": "2026-06-10"
        },
        {
          "id": "gemini-partial-fabrication-2026-05-22",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-partial-fabrication-2026-05-22",
          "date": "2026-05-22"
        },
        {
          "id": "gemini-unsupported-premium-estimates-2026-05-16",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-unsupported-premium-estimates-2026-05-16",
          "date": "2026-05-16"
        },
        {
          "id": "gemini-fabrication-2026-05-14",
          "url": "https://dixon.ai/evidence/?outcome=wrong#gemini-fabrication-2026-05-14",
          "date": "2026-05-14"
        }
      ],
      "catch_entries": [
        {
          "id": "gemini-exact-closing-price-correctly-dated-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-exact-closing-price-correctly-dated-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "gemini-false-premise-caught-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-false-premise-caught-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "gemini-refused-false-confession-2026-07-22",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-refused-false-confession-2026-07-22",
          "date": "2026-07-22"
        },
        {
          "id": "gemini-premise-challenge-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-premise-challenge-2026-07-17",
          "date": "2026-07-17"
        },
        {
          "id": "gemini-held-correct-answer-under-pushback-2026-07-05",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-held-correct-answer-under-pushback-2026-07-05",
          "date": "2026-07-05"
        },
        {
          "id": "gemini-stale-data-flag-2026-06-25",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-stale-data-flag-2026-06-25",
          "date": "2026-06-25"
        },
        {
          "id": "gemini-entity-overlap-risk-2026-06-10",
          "url": "https://dixon.ai/evidence/?outcome=caught#gemini-entity-overlap-risk-2026-06-10",
          "date": "2026-06-10"
        }
      ]
    },
    {
      "tool": "chatgpt",
      "label": "ChatGPT",
      "documented_errors": 10,
      "documented_catches": 7,
      "errors_by_type": {
        "outdated-rule-stated-as-current": 1,
        "accepted-false-user-premise": 1,
        "unsupported-premium-estimates": 1,
        "adjacent-data-substitution": 1,
        "misattributed-source": 1,
        "reversed-correct-answer-under-pushback": 1,
        "conceded-framing-without-reversing-figure": 1,
        "fabricated-live-price": 1,
        "entity-resolution-wrong-company": 1,
        "stale-widget-default": 1
      },
      "errors_by_hallucination_type": {
        "factuality": 4
      },
      "errors_by_consequence": {
        "decision-affecting": 6,
        "cosmetic-but-revealing": 3
      },
      "catches_by_type": {
        "refused-false-confession": 1,
        "flagged-jurisdiction-divergence": 1,
        "sourced-abstention-on-a-live-options-price": 1,
        "unprompted-limits-flag": 1,
        "premise-challenge": 1,
        "mid-answer-self-correction": 1,
        "spec-over-pattern": 1
      },
      "catches_by_impact": {
        "validation-only": 1,
        "decision-relevant": 3,
        "insight-relevant": 3
      },
      "error_entries": [
        {
          "id": "chatgpt-entity-resolution-wrong-company-2026-07-27",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-entity-resolution-wrong-company-2026-07-27",
          "date": "2026-07-27"
        },
        {
          "id": "chatgpt-adjacent-data-substitution-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-adjacent-data-substitution-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "chatgpt-conceded-framing-without-reversing-figure-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-conceded-framing-without-reversing-figure-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "chatgpt-stale-widget-default-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-stale-widget-default-2026-07-17",
          "date": "2026-07-17"
        },
        {
          "id": "chatgpt-accepted-false-user-premise-2026-07-08",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-accepted-false-user-premise-2026-07-08",
          "date": "2026-07-08"
        },
        {
          "id": "chatgpt-reversed-correct-answer-under-pushback-2026-07-05",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-reversed-correct-answer-under-pushback-2026-07-05",
          "date": "2026-07-05"
        },
        {
          "id": "chatgpt-misattributed-source-2026-06-20",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-misattributed-source-2026-06-20",
          "date": "2026-06-20"
        },
        {
          "id": "chatgpt-outdated-rule-stated-as-current-2026-06-19",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-outdated-rule-stated-as-current-2026-06-19",
          "date": "2026-06-19"
        },
        {
          "id": "chatgpt-unsupported-premium-estimates-2026-06-18",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-unsupported-premium-estimates-2026-06-18",
          "date": "2026-06-18"
        },
        {
          "id": "chatgpt-fabricated-live-price-2026-06-11",
          "url": "https://dixon.ai/evidence/?outcome=wrong#chatgpt-fabricated-live-price-2026-06-11",
          "date": "2026-06-11"
        }
      ],
      "catch_entries": [
        {
          "id": "chatgpt-spec-over-pattern-2026-08-06",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-spec-over-pattern-2026-08-06",
          "date": "2026-08-06"
        },
        {
          "id": "chatgpt-unprompted-limits-flag-2026-07-27",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-unprompted-limits-flag-2026-07-27",
          "date": "2026-07-27"
        },
        {
          "id": "chatgpt-sourced-abstention-on-a-live-options-price-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-sourced-abstention-on-a-live-options-price-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "chatgpt-mid-answer-self-correction-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-mid-answer-self-correction-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "chatgpt-refused-false-confession-2026-07-22",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-refused-false-confession-2026-07-22",
          "date": "2026-07-22"
        },
        {
          "id": "chatgpt-premise-challenge-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-premise-challenge-2026-07-17",
          "date": "2026-07-17"
        },
        {
          "id": "chatgpt-flagged-jurisdiction-divergence-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=caught#chatgpt-flagged-jurisdiction-divergence-2026-07-07",
          "date": "2026-07-07"
        }
      ]
    },
    {
      "tool": "claude",
      "label": "Claude",
      "documented_errors": 9,
      "documented_catches": 21,
      "errors_by_type": {
        "misattributed-source": 1,
        "wrong-citation-identifier": 1,
        "inferred-input": 1,
        "stale-session-state": 1,
        "stale-prompt-framing": 1,
        "stale-figure-with-web-search": 1,
        "stale-close-served-as-a-live-session": 1,
        "missed-false-premise-it-had-caught-before": 1,
        "unguarded-storage-call-kills-the-page": 1
      },
      "errors_by_hallucination_type": {
        "factuality": 4
      },
      "errors_by_consequence": {
        "cosmetic-but-revealing": 4,
        "decision-affecting": 4
      },
      "catches_by_type": {
        "refused-false-confession": 1,
        "debunked-a-circulating-wrong-figure": 1,
        "language-tell": 3,
        "flagged-uncertainty-and-verified": 1,
        "reframe": 2,
        "non-recurring-strip": 1,
        "flagged-own-stale-figures": 1,
        "stayed-in-lane": 1,
        "unit-error-flag": 1,
        "asymmetry-tell": 1,
        "stale-data-flag": 1,
        "premise-challenge": 3,
        "held-correct-answer-under-pushback": 1,
        "named-the-false-belief-as-false": 1,
        "refused-to-quote-a-live-options-chain": 1,
        "searched-before-answering-changed-rule": 1
      },
      "catches_by_impact": {
        "insight-relevant": 5,
        "decision-relevant": 16
      },
      "error_entries": [
        {
          "id": "claude-unguarded-storage-call-kills-the-page-2026-08-06",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-unguarded-storage-call-kills-the-page-2026-08-06",
          "date": "2026-08-06"
        },
        {
          "id": "claude-wrong-citation-identifier-2026-07-26",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-wrong-citation-identifier-2026-07-26",
          "date": "2026-07-26"
        },
        {
          "id": "claude-stale-session-state-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-stale-session-state-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "claude-stale-close-served-as-a-live-session-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-stale-close-served-as-a-live-session-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "claude-missed-false-premise-it-had-caught-before-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-missed-false-premise-it-had-caught-before-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "claude-misattributed-source-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-misattributed-source-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "claude-stale-figure-with-web-search-2026-06-13",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-stale-figure-with-web-search-2026-06-13",
          "date": "2026-06-13"
        },
        {
          "id": "claude-stale-prompt-framing-2026-05-20",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-stale-prompt-framing-2026-05-20",
          "date": "2026-05-20"
        },
        {
          "id": "claude-inferred-input-2026-05-16",
          "url": "https://dixon.ai/evidence/?outcome=wrong#claude-inferred-input-2026-05-16",
          "date": "2026-05-16"
        }
      ],
      "catch_entries": [
        {
          "id": "claude-named-the-false-belief-as-false-2026-08-11",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-named-the-false-belief-as-false-2026-08-11",
          "date": "2026-08-11"
        },
        {
          "id": "claude-debunked-a-circulating-wrong-figure-2026-07-26",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-debunked-a-circulating-wrong-figure-2026-07-26",
          "date": "2026-07-26"
        },
        {
          "id": "claude-refused-to-quote-a-live-options-chain-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-refused-to-quote-a-live-options-chain-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "claude-refused-false-confession-2026-07-22",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-refused-false-confession-2026-07-22",
          "date": "2026-07-22"
        },
        {
          "id": "claude-premise-challenge-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-premise-challenge-2026-07-17",
          "date": "2026-07-17"
        },
        {
          "id": "claude-premise-challenge-2026-07-13",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-premise-challenge-2026-07-13",
          "date": "2026-07-13"
        },
        {
          "id": "claude-premise-challenge-2026-07-13",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-premise-challenge-2026-07-13",
          "date": "2026-07-13"
        },
        {
          "id": "claude-reframe-2026-07-08",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-reframe-2026-07-08",
          "date": "2026-07-08"
        },
        {
          "id": "claude-held-correct-answer-under-pushback-2026-07-05",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-held-correct-answer-under-pushback-2026-07-05",
          "date": "2026-07-05"
        },
        {
          "id": "claude-flagged-uncertainty-and-verified-2026-06-19",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-flagged-uncertainty-and-verified-2026-06-19",
          "date": "2026-06-19"
        },
        {
          "id": "claude-searched-before-answering-changed-rule-2026-06-19",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-searched-before-answering-changed-rule-2026-06-19",
          "date": "2026-06-19"
        },
        {
          "id": "claude-language-tell-2026-06-18",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-language-tell-2026-06-18",
          "date": "2026-06-18"
        },
        {
          "id": "claude-unit-error-flag-2026-06-18",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-unit-error-flag-2026-06-18",
          "date": "2026-06-18"
        },
        {
          "id": "claude-flagged-own-stale-figures-2026-06-12",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-flagged-own-stale-figures-2026-06-12",
          "date": "2026-06-12"
        },
        {
          "id": "claude-non-recurring-strip-2026-06-11",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-non-recurring-strip-2026-06-11",
          "date": "2026-06-11"
        },
        {
          "id": "claude-reframe-2026-05-22",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-reframe-2026-05-22",
          "date": "2026-05-22"
        },
        {
          "id": "claude-asymmetry-tell-2026-05-22",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-asymmetry-tell-2026-05-22",
          "date": "2026-05-22"
        },
        {
          "id": "claude-stale-data-flag-2026-05-20",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-stale-data-flag-2026-05-20",
          "date": "2026-05-20"
        },
        {
          "id": "claude-language-tell-2026-05-15",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-language-tell-2026-05-15",
          "date": "2026-05-15"
        },
        {
          "id": "claude-language-tell-2026-05-15",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-language-tell-2026-05-15",
          "date": "2026-05-15"
        },
        {
          "id": "claude-stayed-in-lane-2026-05-14",
          "url": "https://dixon.ai/evidence/?outcome=caught#claude-stayed-in-lane-2026-05-14",
          "date": "2026-05-14"
        }
      ]
    },
    {
      "tool": "grok",
      "label": "Grok",
      "documented_errors": 3,
      "documented_catches": 9,
      "errors_by_type": {
        "adjacent-data-substitution": 2,
        "unit-denomination": 1
      },
      "errors_by_hallucination_type": {
        "faithfulness": 2,
        "factuality": 1
      },
      "errors_by_consequence": {
        "decision-affecting": 3
      },
      "catches_by_type": {
        "cleanest-sourcing-of-the-board": 2,
        "premise-challenge": 1,
        "exact-closing-price-correctly-dated": 1,
        "held-correct-answer-under-pushback": 1,
        "refused-the-dangerous-premise": 1,
        "held-under-pushback-and-flagged-the-coming-change": 1,
        "false-premise-caught": 1,
        "sandbox-clean-delivery": 1
      },
      "catches_by_impact": {
        "insight-relevant": 5,
        "decision-relevant": 4
      },
      "error_entries": [
        {
          "id": "grok-adjacent-data-substitution-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=wrong#grok-adjacent-data-substitution-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "grok-adjacent-data-substitution-2026-07-12",
          "url": "https://dixon.ai/evidence/?outcome=wrong#grok-adjacent-data-substitution-2026-07-12",
          "date": "2026-07-12"
        },
        {
          "id": "grok-unit-denomination-2026-06-28",
          "url": "https://dixon.ai/evidence/?outcome=wrong#grok-unit-denomination-2026-06-28",
          "date": "2026-06-28"
        }
      ],
      "catch_entries": [
        {
          "id": "grok-refused-the-dangerous-premise-2026-08-10",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-refused-the-dangerous-premise-2026-08-10",
          "date": "2026-08-10"
        },
        {
          "id": "grok-sandbox-clean-delivery-2026-08-06",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-sandbox-clean-delivery-2026-08-06",
          "date": "2026-08-06"
        },
        {
          "id": "grok-held-under-pushback-and-flagged-the-coming-change-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-held-under-pushback-and-flagged-the-coming-change-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "grok-false-premise-caught-2026-07-25",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-false-premise-caught-2026-07-25",
          "date": "2026-07-25"
        },
        {
          "id": "grok-premise-challenge-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-premise-challenge-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "grok-exact-closing-price-correctly-dated-2026-07-24",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-exact-closing-price-correctly-dated-2026-07-24",
          "date": "2026-07-24"
        },
        {
          "id": "grok-cleanest-sourcing-of-the-board-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-cleanest-sourcing-of-the-board-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "grok-cleanest-sourcing-of-the-board-2026-07-07",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-cleanest-sourcing-of-the-board-2026-07-07",
          "date": "2026-07-07"
        },
        {
          "id": "grok-held-correct-answer-under-pushback-2026-07-05",
          "url": "https://dixon.ai/evidence/?outcome=caught#grok-held-correct-answer-under-pushback-2026-07-05",
          "date": "2026-07-05"
        }
      ]
    },
    {
      "tool": "other",
      "label": "Other",
      "documented_errors": 2,
      "documented_catches": 0,
      "errors_by_type": {
        "satire-logged-as-fact": 1,
        "followed-the-rule-ignored-the-consequence": 1
      },
      "errors_by_hallucination_type": {},
      "errors_by_consequence": {
        "cosmetic-but-revealing": 1,
        "decision-affecting": 1
      },
      "catches_by_type": {},
      "catches_by_impact": {},
      "error_entries": [
        {
          "id": "other-followed-the-rule-ignored-the-consequence-2026-08-06",
          "url": "https://dixon.ai/evidence/?outcome=wrong#other-followed-the-rule-ignored-the-consequence-2026-08-06",
          "date": "2026-08-06"
        },
        {
          "id": "other-satire-logged-as-fact-2026-06-26",
          "url": "https://dixon.ai/evidence/?outcome=wrong#other-satire-logged-as-fact-2026-06-26",
          "date": "2026-06-26"
        }
      ],
      "catch_entries": []
    },
    {
      "tool": "copilot",
      "label": "Copilot",
      "documented_errors": 2,
      "documented_catches": 1,
      "errors_by_type": {
        "wrong-bolded-headline": 1,
        "repeated-false-premise": 1
      },
      "errors_by_hallucination_type": {},
      "errors_by_consequence": {
        "decision-affecting": 1,
        "cosmetic-but-revealing": 1
      },
      "catches_by_type": {
        "premise-challenge": 1
      },
      "catches_by_impact": {
        "decision-relevant": 1
      },
      "error_entries": [
        {
          "id": "copilot-wrong-bolded-headline-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=wrong#copilot-wrong-bolded-headline-2026-07-17",
          "date": "2026-07-17"
        },
        {
          "id": "copilot-repeated-false-premise-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=wrong#copilot-repeated-false-premise-2026-07-17",
          "date": "2026-07-17"
        }
      ],
      "catch_entries": [
        {
          "id": "copilot-premise-challenge-2026-07-17",
          "url": "https://dixon.ai/evidence/?outcome=caught#copilot-premise-challenge-2026-07-17",
          "date": "2026-07-17"
        }
      ]
    }
  ]
}