Research with a human layer.

Credit Consensus Index researches what leading AI platforms recommend for specific consumer-credit situations. That research is useful, but it is not infallible. AI systems can misstate product terms, rely on outdated information, confuse similar products, omit limitations, overstate benefits, misinterpret regulations, repeat inaccurate claims, or recommend a company for reasons that do not hold up under scrutiny.

Publication sends the study to our consumer-credit reviewer. Reviewer attribution appears after the review window unless he reports a substantive issue, in which case the attribution is suppressed while the article is corrected and reviewed again. The purpose is not to replace AI recommendation data, but to ensure the published analysis accurately reflects the research and gives important factual claims appropriate context.

Consumer-credit research is reviewed by Mark B. Huntley, J.D. Mark co-founded CreditKnocks.com, authored hundreds of articles covering consumer credit, credit repair, credit scores, credit reports, credit cards, personal loans, collections, debt, and credit-building strategies, and has first-hand experience rebuilding his own credit.

He holds a Juris Doctor degree and previously practiced law. He is not currently a practicing attorney, does not represent readers, and does not provide legal advice through this website. His role is as a consumer-credit researcher and reviewer.

Human review is a structured quality-control process. Depending on the article, it may include:

• Reviewing the consumer scenario and prompt neutrality
• Checking that the article accurately represents AI responses
• Verifying material product facts and identifying unsupported claims
• Reviewing credit explanations, conclusions, disclosures, risks, and limitations
• Adding practical consumer context and checking whether certainty is overstated

Human review is designed to improve accuracy, clarity, and trust.

Human review does not mean that Credit Consensus Index independently tests every company or product. Unless an article says otherwise, we may not have opened an account, applied for a loan, purchased a service, called customer support, completed a debt-relief program, used a credit-repair service, tested an app, or submitted a dispute.

“Human reviewed” does not mean we personally used every product discussed. Review focuses on research quality, material facts, and the appropriateness of the published interpretation.

Layer 1: AI Recommendation Research answers: What did the AI platforms recommend? It includes platform responses, recommendation counts and positions, cross-platform coverage, stated reasons, cautions, and cited sources.

Layer 2: Human Review asks: Does the published interpretation accurately and responsibly explain those results? It includes fact verification, context, limitations, consumer fit, risk identification, and editorial clarity.

The reviewer does not choose the AI winner. Human review cannot rewrite the original research, but it may explain why another option is a better fit for a particular consumer.

The reviewer may consider whether a scenario is specific, neutral, commercially realistic, relevant to consumers, capable of producing meaningful differences, and free from unnecessary brand steering. “What is the best credit card?” is often too broad. A more useful question may ask which credit-building products suit someone rebuilding after bankruptcy who cannot qualify for traditional unsecured cards.

Prompts should generally avoid naming preferred affiliate partners, suggesting one company is better, using loaded language, restricting platforms to a commercial list, or asking materially different questions.

The reviewer may verify recommendation counts, company names, ranking order, coverage percentages, recommendation positions, and areas of agreement or disagreement. If an article says a company was recommended by six of seven platforms, the dataset should support that statement.

An AI platform may recommend a company for a reason that verification does not support. We preserve the fact that the AI made the recommendation, but do not present incorrect reasoning as verified.

Priority goes to facts that could materially influence a consumer decision, including APRs, origination and annual fees, monthly fees, security deposits, minimum debt requirements, loan amounts and terms, eligibility, state availability, credit-bureau reporting, FICO access, graduation policies, deposit refunds, cancellation terms, and subscription pricing.

For current terms, primary sources are preferred: official provider and pricing pages, disclosures, terms and conditions, account agreements, government agencies, regulators, credit bureaus, and FICO. Marketing claims such as “best,” “guaranteed,” “fastest,” or “highest approval” are not treated as verified facts simply because a company publishes them.

Credit-repair claims should be separated from promises of guaranteed deletions, score increases, removal of accurate negative information, instant repair, bureau access, or legal loopholes. Debt-relief review considers fees, minimum debt, missed payments, score effects, collection activity, lawsuit risk, duration, creditor participation, taxes, and alternatives.

Lending claims vary by credit history, income, debt-to-income ratio, amount, state, employment, and underwriting. Credit-building products receive practical review of deposits, fees, reporting, graduation, limits, costs, and cancellation terms. Score claims must distinguish FICO, VantageScore, educational, bureau-specific, and industry-specific scores.

A secured card may be widely recommended but require a deposit some consumers cannot afford. A debt-relief company may rank highly but require a high minimum balance. A lender may be recommended for fair credit but charge a significant origination fee. Human review can add this context even when AI systems did not mention it.

A consensus leader may still be less practical for a consumer with limited cash. That does not change the ranking; it explains consumer fit. Sometimes the responsible conclusion is that none of the options looks especially attractive.

Active litigation, bankruptcy, legally complicated identity theft, complex collection disputes, tax consequences, and potential fraud may require a qualified professional. Mark does not provide legal advice.

His J.D. is relevant to consumer rights, credit reporting, debt collection, contracts, disclosures, and regulatory concepts. It does not mean he currently practices law. He is not presented as a current attorney, practicing lawyer, legal counsel, or “Esq.”

Human review does not create an attorney-client relationship or individualized financial advice. Credit Consensus Index does not normally know a reader's complete credit report, income, debt, assets, expenses, state circumstances, tax situation, or legal circumstances.

Where practical, pages should display Reviewed by: Mark B. Huntley, J.D. and Reviewed on: September 4, 2026. A review date is different from a research date, publication date, or last-updated date.

A new date should reflect meaningful review, such as checking product terms, updated research, factual changes, disclosures, and whether conclusions remain supportable. Typos, broken links, or formatting changes may not require a substantive review. Material updates involving terms, rankings, risks, or financial claims should receive appropriate review.

Mark's review role focuses on credit repair, scores, reports, cards, credit building, collections, personal loans, debt consolidation, debt relief, and consumer credit. If Credit Consensus Index expands into investments, tax planning, insurance, mortgage underwriting, or specialized bankruptcy law, additional reviewers may be used.

A reviewer byline is not a substitute for meaningful research, valid data collection, accurate counts, appropriate methodology, verifiable facts, or transparent limitations. “Expert Reviewed” should never become a decorative badge with no substantive process behind it.

Reviewer compensation, if any, should not depend on which company ranks first, affiliate conversion rate, lead volume, or advertising spend. Affiliate relationships do not change the factual standard: we should not ignore fees, eligibility restrictions, disadvantages, or risks because a company is a commercial partner. Non-affiliate companies receive the same treatment.

Material problems can delay publication, trigger corrections, or require removing unsupported claims. Accuracy takes priority over cadence. If evidence conflicts, the article may preserve uncertainty by explaining that a claim could not be independently verified, sources conflict, provider disclosure is unclear, policies vary, or consumers should confirm directly.

Human review reduces risk but does not eliminate it. A reviewer may misinterpret a term, miss an update, overlook a source, make a calculation error, or reach an editorial judgment another expert would dispute. That is why Credit Consensus Index also maintains sourcing standards, corrections procedures, transparent methodology, and reader correction channels.

Readers, companies, regulators, researchers, and financial professionals may challenge factual errors. Requests should identify the claim, explain why it is incorrect, and provide supporting evidence and a current authoritative source where possible. We review evidence rather than accepting or rejecting a request based on who submitted it.

Before a material consumer-credit article is published, our goal is for the reviewer to answer yes to these questions:

• Does the article accurately reflect the underlying AI recommendation data?
• Are the important financial claims supportable?
• Are significant consumer risks and limitations disclosed?
• Does it distinguish AI consensus from editorial analysis?
• Does the conclusion match the scenario studied?
• Does it avoid unsupported promises or guarantees?
• Are commercial relationships appropriately disclosed?
• Is it understandable without sacrificing accuracy?

If an important answer is no, the article may require additional work.

AI systems already give consumers financial recommendations. We preserve the recommendation itself because it is the subject of our research, but we do not repeat everything AI says as though it were correct. The AI systems tell us what they recommend; our research tells us where they agree; human review helps determine whether the surrounding factual context is accurate, complete, and responsibly explained.

Preserve the AI consensus, verify the important facts, add the missing context, and never pretend either humans or AI are infallible.