EVIDENCE & RESEARCH

A claim should have to earn support.

Argumont separates what is being claimed from what actually supports the claim. Research exists to change the Decision when the facts justify it, not to decorate an answer with links.

Research follows materiality.

Argumont should ask: if this fact changed, could it materially change the Verdict?

If the answer is no, it usually does not deserve the same research effort as a load-bearing claim.

Research depth should follow the stakes of the Decision and the importance of the unknown.

Scout investigates.

Scout gathers outside intelligence relevant to the Decision.

Depending on the domain, that can include official records, regulatory information, technical documentation, current pricing, product specifications, company filings, research, market information, reputable testing, public datasets, credible reporting, and relevant user experience where appropriate.

Scout's job is to find the information.

Scout does not get to decide that a source proves more than it actually proves.

Inspector inspects the source.

Inspector asks whether the source is authoritative for this claim, whether it is current enough, whether it directly supports the statement, whether it applies to the exact product, year, jurisdiction, company, population, configuration, or situation being discussed, whether the claim is broader than the evidence, whether there is contradictory evidence, and what remains unresolved.

Argumont uses Verified only when an authoritative government or public-authority source directly confirms the exact claim. Company, platform, academic, news, secondary, and community sources may support or qualify a claim. A link is not verification.

Source quality depends on the domain.

For automotive questions, official safety and recall information may matter more than a random forum comment.

For legal questions, statutes, regulations, courts, agencies, and official bar authorities may matter more than a blog summary.

For cybersecurity questions, official vendor advisories, CISA, vulnerability records, technical documentation, and credible security research may matter more than generic articles.

For financial questions, filings, regulators, primary company information, and reliable market data may matter more than promotional content.

Argumont should adapt its evidence hierarchy to the Decision.

Search against the emerging favorite.

Research should not become confirmation search.

When one option begins to lead, Argumont should actively look for material evidence that could weaken the leading case.

A strong Verdict should survive evidence that was not selected merely because it was convenient.

Preserve scope.

If an official source establishes a documented failure risk affecting a defined population, Argumont must not transform that into a claim that every unit will fail.

If a study describes one population, Argumont must not silently generalize it to everyone.

If a regulation applies in one jurisdiction, Argumont must not present it as universal.

Inspector exists partly to prevent this type of overclaim.

Unknown is a valid result.

When reliable evidence cannot resolve a material fact, Argumont should say so.

Unknown, unverified, disputed, estimated, and assumed are more useful than false certainty.

Bring Argumont a Decision worth challenging.

Start with a paid 60 day pilot on your own real Decisions.

Research the Decision that matters.