Answer once. Get paid. Teach the model.

Approved legal professionals, engaged per Work Request and paid per answer, not per hour. Every question you answer is measured and paid. Where your answer becomes knowledge the system reuses, you share in what FinePrint LM earns for as long as it keeps working.

You keep your practice. This is a queue, not a job.

Work Requests come to your queue through AnswerLoop, scoped to your state and practice area, with the file already built. You accept work on standards you can read before you accept anything. It is not a referral scheme and it is not piecework.

01

Paid per Work Request

A published rate per Work Request, from $20 to $500 and more, set by the request type before you accept it. It is the same whether you agree, modify or rework. Pay never depends on the answer.

02

Revenue share on knowledge

Contribute a playbook, a position, a clause pattern. When endpoints use it, you share the revenue it earns, for as long as it keeps working. Contribution stays attributed.

03

Your answer trains the model

Every answer you give is what FinePrint LM is made of. The standards you mark to are written and versioned. The product cannot override a flag you raise.

Matter work is separate. When a customer’s matter goes Yellow or Red, a conflict-checked attorney is engaged by the customer, on a stated scope, with the file already built: the record, the documents, the history and the issue list. Who reviews the work

What we ask of a reviewer.

  • Licensure in good standing. In the jurisdictions whose queues you accept, re-verified on a schedule.
  • Professional liability coverage where an answer clears a customer’s matter. Labeling work that only trains the model does not need it.
  • Conflicts, disclosed and re-checked. Before a matter is offered, and again when the facts change.
  • A turnaround you agreed to. Each queue states its window. You accept the queue, not an open-ended obligation.
  • Structured feedback, not free text. Agree, modify or rework, and name the governing reason: the statute, the standard, or the practice that decides it.
  • Confidentiality on anonymized material. Contributor terms bind everything you see in sampled review, including a bar on re-identification.

What attorneys ask us.

What happens to work I contribute?

It arrives with provenance, consent and a permitted use, and it stays attributed. You are paid per Work Request. You share in the revenue the knowledge you contributed earns, for as long as it keeps working. Your answers train the model. Your pay never depends on the answer.

Could the rate pressure me toward approving?

No. The rate per Work Request is set by the request type before you accept it and is the same for an accept, a correction, a rejection and a stop. There is no throughput bonus tied to outcomes. Revenue share attaches to contributed knowledge, not to answers. A reviewer who flags often earns the same as one who flags rarely, for the same volume.

What do I see in sampled review?

Anonymized artifacts: the document structure, jurisdiction, legal posture and clause language intact, with the company, the people, the counterparties and identifying amounts removed. You never learn whose work it is, and the terms bar you from trying. A live customer matter is different. You see it only when that customer has engaged you on it.

Does this create an attorney-client relationship?

Sampled review does not. You are reviewing anonymized artifacts under contributor terms, not advising a client. A matter engagement does. It runs between you and the customer directly, conflict-checked, on a stated scope. Ask us which obligations attach in your jurisdiction before you accept a queue. The standards and terms spell out the structure.

Tell us your jurisdictions. We’ll send the standards.

The review standards for the queues that fit your practice, and the reviewer terms. In writing, before you commit to anything. We’ll get back to you as soon as possible.

We use these details only to send you the review standards and talk about the network. Nothing else, and no mailing list. See our Privacy Policy.