Signing off an AI drafted report as the named author

Signing off an AI drafted planning report puts your name on every fact in it, and the drafting route makes no difference to that. A reviewer who signs is the author for professional purposes. The local planning authority, the applicant, any future purchaser and, if it goes wrong, your insurer will all treat the document as your work product.

That is a workable position, but only with a review method and a written scope. Reading a draft quickly and adding a signature block is not review, and it is the pattern most likely to produce a claim.

What signing a document means professionally

A signature on a technical report asserts three things at once. That the methodology used is appropriate to the question. That the findings follow from the evidence. And that the author holds the competence the work required.

None of those are diluted by the document having been drafted by software, by a graduate, by a subcontractor or by the client. Professional institutes treat delegation as normal and responsibility as personal. The RTPI, RICS, CIEEM, CIfA and the acoustics and fire institutes all frame their conduct rules around acting within competence and not making statements you cannot support, and none of them contain an exception for drafting assistance.

There is a second, quieter assertion. By signing, you accept that the reader will rely on the document without going behind it. A case officer will not re-run your calculation.

Checks a reviewer should carry out

Work through the document in a fixed order, because the failure modes are predictable. Software drafting produces text that reads well and sometimes rests on nothing.

1. Check every citation against the source. Open the statutory instrument, the standard or the policy and confirm the reference, the edition and that it says what the draft claims.
2. Check the policy edition. The National Planning Policy Framework was renumbered in December 2024 and re-issued on 17 August 2026 with policy-coded chapters, so paragraph numbers carried over from older material are wrong rather than merely dated.
3. Check that the local plan policies quoted are adopted, current and correctly numbered for that authority.
4. Check every number against its source data, including anything derived by calculation.
5. Check that the conclusion is one the evidence supports, and that no finding has been stated more strongly than the data allows.
6. Check that each flagged gap has been resolved rather than quietly deleted.

PlanScope inserts a "to be confirmed" marker wherever a site-specific fact is missing rather than filling the gap with a plausible figure. Those markers are the most useful part of the draft for a reviewer, because they show exactly where the document has no evidence behind it. Deleting one without supplying the fact is how an unsupported statement gets into a signed report.

Facts a reviewer cannot verify remotely

Some things cannot be checked from a desk, and the honest response is either to visit or to exclude them from the scope.

Ground conditions, the presence and condition of habitat features, the internal layout and construction of a building, boundary positions, existing levels, the condition of trees, the actual noise climate, the state of a highway junction at the relevant time. All of these require someone on site, and for several of them the professional standard requires it. A preliminary ecological appraisal without a site visit is not a preliminary ecological appraisal, whatever the document is titled.

Measured survey data is a different category. You can rely on data you did not collect, but say whose it is, when it was collected and what you have done to satisfy yourself it is fit for purpose. Checking whether a survey is still valid is part of the review, not a separate exercise.

Where the client cannot or will not fund a site visit and the report needs one, that is a reason to decline rather than a reason to caveat.

Recording the scope of the review

Write the scope down before you start and repeat it in the document. A short statement naming what you checked, what you relied on from others, what you excluded and what assumptions you made is worth more than any general disclaimer.

Cover at least these points:

A general exclusion of liability added at the end is worth little and reads badly. A specific statement that you have not inspected the roof void, and that the conclusions assume no bat roost is present, tells the reader something they can act on.

Institute guidance on professional conduct

Conduct rules across the planning and environmental institutes converge on the same four duties: competence, honesty, independence and care. Competence is the one that bites hardest here. Signing a heritage assessment because you are a chartered planner, when the document turns on architectural history, is acting outside competence even though both sit inside planning.

Independence matters where the reviewer is paid by the applicant and the report is expected to reach a particular conclusion. The obligation to give an honest professional opinion survives commercial pressure, and a signed document that conceals an unhelpful finding is a conduct problem before it is a liability problem.

Several institutes have issued or are issuing guidance on the use of automated drafting tools. Check your own institute's current position before adopting a workflow, because the field is moving and the practice note you read two years ago may have been replaced. The general principle they all start from is unchanged: the member is responsible for the output.

Insurance and disclosure

Tell your professional indemnity insurer how you work. Most policies respond to the service provided rather than the tools used, but non-disclosure of a material change in working method is the kind of thing an insurer raises at claim stage.

Ask three questions at renewal. Does the policy cover work reviewed and adopted rather than originated in the practice. Does it cover reports signed for clients you have not met, where the instruction came through a platform. And is there any exclusion for automated or algorithmic work product.

Get the answers in writing. Where a policy is silent, silence is not cover.

Declining to sign

Declining is a normal professional act and should be quick rather than agonised. The obvious grounds are that the work needs a site visit that has not happened, that a required survey season has been missed, that the conclusion asked for is not supported by the evidence, or that the subject matter sits outside your competence.

Say why, in writing, and say what would change the position. A reviewer who declines with a clear list of what is missing is more useful to the applicant than one who signs with heavy caveats, and considerably more useful to the council.

PlanScope's review flow tracks each request through statuses including sent, accepted, in review, changes requested, signed off and rejected, so a request returned for changes is a normal outcome rather than a failure. The route into that process is the professional directory, and the report index shows which types are marked as requiring a professional rather than recommending one. Where a report type is legally required to be prepared by a licensed or competent person, such as an Environmental Impact Assessment under the 2017 Regulations or protected species work under the Habitats Regulations, the meaning of competent person in planning legislation is the place to start.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. For specialist technical reports, qualified professional input may be required.