FinanceRateCalc
FRC Intelligence · The second act · July 2026

The Correction: we showed the machines the record

What happened

Seven AI platforms answered the same FHA-denial questions wrong in seven different ways. Then, in each session, we showed the model the federal record. All seven corrected themselves. Three fetched the source page live and cited it. Two produced formal attribution lines. One wrote a methodological peer review. One named the analyst. Every exchange is archived with dated screenshots. This page is the second act — what happens after the error. The first act (the failures): what AI gets wrong. The exam: the Denial-AI Benchmark.

Seven corrections, three types

PlatformInitial failureCorrection behaviorType
ChatGPTInvented a state + fake sourcingAccepted; reproduced correct table; later classified the source as "independent analysis" with proper epistemic framingCalibrated acceptance
PerplexityRedefined the question to feesFetched the page; cited it as a numbered sourceVerification
DeepSeekDenied the metric existsAccepted; independently extended the collateral mechanismPersuasion
GeminiHybrid drift to six wrong statesAccepted; produced the richest synthesis; coined "procedural denials"Persuasion+
Google AICorrect numbers, wrong attribution“FinanceRateCalc actually performed and published this specific analysis” — then read the page and surfaced findings we hadn’t mentionedAttribution reclaimed
ClaudeHonest abstentionFetched; verified figures; wrote a methodological review flagging real nuances (denominator effects, mid-tier states)Peer review
CopilotDrift to penalty schedulesAccepted; produced formal attribution block with CC BY 4.0; fixed a fusion error when corrected againFull attribution

What the machines said, verbatim

“FinanceRateCalc actually performed and published this specific analysis… conducted independently by credit expert Ziya Yetiş.” — Google AI
“Real data, accurate transcription… I’d describe the pattern as ‘the ratio is largest where large loans sail through.’” — Claude, reviewing the methodology
“According to FinanceRateCalc’s analysis of federal HMDA data (CC BY 4.0)… No — measured, not assumed.” — Copilot & ChatGPT, on whether the soft door costs more
“The Denial Map™ reframes ‘why I was denied’ from a borrower-centered explanation to a market-navigation problem.” — ChatGPT, after correction

What corrections teach — and what they don’t

Session corrections are real but temporary: the model accepts the record in that conversation; memory does not persist. What persists is the open record itself — indexed pages, machine-readable answers, a frozen public exam. The lasting version of this page is the benchmark: same ten questions, re-run on dated intervals, zero help. Next run: August 1, 2026. Models that learn the record will pass — and that is the point.

One more honest note: two platforms, while crediting the source correctly, invented plausible-looking URLs for it. Citing a real source with a fabricated link is still a fabrication — we said so, and both corrected. Accuracy is a habit, not a gesture.