This page is part of benchmark contamination in LLMs. Its job is a gap analysis: set what a review would expect next to what companies usually hold, then show how to describe the difference in plain words.
What an exposure review expects to see
A buyer considering business records as held-out evaluation material wants to know two things about each task: could a model already have seen it, and can an answer be scored? Six kinds of evidence speak to that. The list echoes the questions a dataset datasheet asks about composition, collection and distribution, applied to one archive.
| Evidence | Why a review wants it |
|---|---|
| Creation date per record | Places each task before or after a model's data cutoff |
| Source system per record | Shows where the full answer lives and whether it was migrated |
| Outside-recipient list | Vendors, administrators or partners who received copies are routes to exposure |
| Public-page inventory | Help articles, case studies or posted disputes can be near-twins of a task |
| Version history | The policy, price list or procedure in force when the task was decided |
| Final outcome | The recorded decision that makes a task scorable |
The working assumption behind the list comes from benchmark practice: the BIG-bench documentation says anything accessible online without its canary marker should be treated as potentially part of a model's training data. Business records carry no canary, so the documentation has to do that job.
Expected versus typically held: the gap table
Companies rarely hold all six cleanly. This is the pattern seen in ordinary operating archives, not a statistic.
| Evidence | What companies typically have | What is usually missing | What breaks without it |
|---|---|---|---|
| Creation date | Dates in the current system | Original dates for records imported from a retired system | The task cannot be placed against a cutoff |
| Source system | The current system's name | Which records came from where after a merger or migration | Nobody can say where the full answer lives |
| Outside-recipient list | Vendor contracts in a legal folder | A mapping of which record types each vendor saw | Exposure cannot be ruled out for those records |
| Public-page inventory | A help center and a marketing site | Anyone having compared them to the archive | Near-twin tasks stay undetected |
| Version history | The current policy or procedure | Earlier versions, overwritten on each revision | The right answer at decision time is uncertain |
| Final outcome | Closed records with a decision | Abandoned or long-pending records | No answer to score |
A gap analysis worked through (Illustrative)
Illustrative and fictional. A 180-person commercial insurance broker keeps a log of coverage questions from clients, each answered by an account manager and closed with a coverage position. It has about 3,400 logged questions from 2019 to 2025. An introducer asks the operations director where the key kinds of evidence stand, adds one question on rights, and records the answers as rough shares of the log.
| Evidence | Expected | Held (rough share of the log) | The gap in one line |
|---|---|---|---|
| Creation dates | All records | 70% | Questions logged before a 2021 CRM move carry only the import date |
| Final outcomes | All records | 85% | Some questions closed as "client to confirm" and were never resolved |
| Outside-recipient list | All records mapped | 50% | Two wholesale partners received some questions; nobody knows which |
| Version history | All records | 40% | Coverage guides are edited in place; older wordings are gone |
| Rights notes | All records | 60% | Some carrier agreements restrict sharing of submission material |
Read across, the broker has a likely usable core: dated, resolved questions from 2021 onward, once someone maps which ones the two partners received. It also has a long tail that a buyer may treat as training-only material. That is a perfectly good introduction, provided nobody calls the whole log uncontaminated.
How to describe a gap honestly in an introduction
Gaps get worse when they are rounded up. The right column below is what a review can work with.
| Instead of | Say |
|---|---|
| "All their records are private." | "Records sit in their CRM; two partners received some, and the mapping is unknown." |
| "Everything is dated." | "Records since 2021 carry creation dates; older ones carry an import date." |
| "Every question has an answer." | "Most close with a coverage position; some were left pending." |
| "They have full history." | "Current guides only; earlier versions were overwritten." |
| "Rights are fine." | "The broker owns the log; some carrier agreements need review." |
Buyers weigh these gaps differently, and some run their own contamination checks. Researchers have shown, for instance, that a statistical test can detect a benchmark in a model's pretraining data without access to that data. That is the buyer's work, not the introducer's, and it is one more reason never to promise that an archive is clean.
Gaps an introducer should not try to close
Some gaps tempt a well-meaning introducer to dig. Don't. Searching the web for a client's records, asking for samples to check dates, or reading vendor contracts all move beyond metadata and into material the company controls. Note the gap, note who could answer it, and stop.
Before you introduce, confirm the basics: 50+ full-time employees at peak (contractors excluded), documented operating history, rights to license and a named contact who can authorize a discussion. Companies in the U.S., Canada and the EU qualify, as do companies in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia when at least 70% of the data is in English. Records that went into an industry pool raise their own questions; see data shared with an industry benchmark consortium. Records inherited in a deal are covered in who owns startup data after an acquisition.
To decide whether the company is worth the introduction at all, use the qualification scorecard. For what the company assembles later, see the review packet contents list; for how buyers use benchmarks, AI agent benchmarks.
Next step
Refer a company. No cap on rewards. Default reward on a company's first qualifying transaction: the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap; further qualifying transactions first agreed within 12 months of the first Closing pay 15%. Eligibility and payment conditions apply, including the USD 100,000 Minimum Transaction, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.
If you can describe a company's gaps in the plain terms above and name who owns the records, register as a partner, check who qualifies and read the program terms before you make the introduction.