Scope the AI task before anyone scopes your file transfer.

Start here

Does any of this sound familiar?

• A potential partner asks for examples but has not clearly described the intended model task.
• Your proposals contain repeatable diagnostic or planning workflows mixed with client-specific facts.
• You want a disciplined way to assess a request before agreeing to a sample, pilot, or transfer.

A proposal request may seek task benchmarking, model training, feature evaluation, or historical records. Each use has different evidence and rights needs.

Start with the task and criteria, not a promise to provide documents. A brief can assess fit without revealing a proposal, client, pricing, or confidential method.

The problem

Vague scope makes every safeguard harder to test

A requester wants 'consulting proposals' for AI. Does it need discovery-question checks, a comparison against constraints, or an engagement plan? Is it training, testing, or asking experts to create tasks? Without scope, the firm cannot set inputs, scoring, or disclosure rights.

Proposals may reveal client context, staff, budgets, risks, pricing, and methods. Removing a name may not anonymize them. A bid or pilot discussion is not a closed license; record status so exploration is not mistaken for approval.

If the task cannot be stated without asking for a client file, the scope is not ready.

The solution

Use a five-question scope brief

A written brief lets legal, delivery, security, and commercial owners proceed, revise, or decline.

DataSupply partners only with labs that meet its top 0.01% credibility standard. We help assess whether a qualified buyer may be a fit and negotiate terms that reflect the data's potential value, including exclusivity where relevant. We also help you work through diligence questions about rights, privacy, security, and compliance, then present a high-level inventory of permitted records, not the dataset. Fit is specific to each situation; no buyer or value is guaranteed.

What to inventory before any buyer conversation

• Name the user and work decision Ask who uses the system, which consulting task it supports, what decision follows, and whether an expert remains responsible. Replace 'strategy data' with a task such as checking plan dependencies.
• Specify the minimum evidence and use Clarify whether the partner needs authored tasks, evaluation examples, licensed history, or a mix. Define training or testing, output use, review, format, and whether synthetic cases suffice.
• Write the boundary and decision trail Document permitted inputs, exclusions, rights owner, access, transfer, retention, deletion, derivative handling, security contact, and exit criteria. Name the decision owner and status.

Set the boundaries before discussing access.

Verify the counterparty. Counsel should review confidentiality, IP, privacy, subcontracting, training, onward use, audit, and breach terms. Use least privilege and secure transfer only after approval. An NDA does not grant missing rights.

What could make a permitted example useful?

A precise scope may favor task creation over historical records. Compare expert time, clearance, review, security, and strategic risk; a bid is not a completed agreement or guaranteed result.

A practical first step.

Ask the requester for a scope brief before sharing examples. Review it without client files and proceed only when purpose, rights, and controls are clear.

datasupply.ai can discuss possible fit and buyer questions without receiving your dataset. You decide whether to pursue any introduction. No buyer, license, or payment is guaranteed.

Documented example / what it proves

Public procurement guidance starts with a defined acquisition decision

UK Government Digital Service guidance helps public-sector buyers evaluate suppliers and procure AI responsibly. It is analogous evidence for treating AI requests as scoped evaluation and acquisition, not an open-ended request for supplier data. Read UK Government Digital Service, Guidelines for AI procurement.

It concerns government procurement and grants no private firm's rights, client consent, or model-training terms. Use it to prompt clear task, supplier role, evaluation, and accountability questions; then review contracts.

The important limit: This public guidance is not a closed consulting-data license, completed bid, buyer commitment, or client permission.