> For the complete documentation index, see [llms.txt](https://docs.filed.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.filed.com/reference/products/ai-review/frequently-asked-questions.md). # FAQ ## What does AI Review actually check? AI Review cross-checks the values on a draft return against the source documents in the binder: W-2s, 1099s, K-1s, payroll summaries, prior-year returns, trial balances, and similar materials. It catches value mismatches between what the draft shows and what the evidence supports, missing items flagged in the binder, and form-level inconsistencies. AI Review checks the return against the evidence in the binder; it does not evaluate compliance rules in isolation. ## Which return types does AI Review support? AI Review supports: Form 1040 (individuals), Form 1120-S (S corporations), Form 1065 (partnerships), Form 1120 (C corporations), Form 1041 (estates and trusts), and Form 990 (tax-exempt organizations). Checks are scoped automatically to the forms applicable to the client's configured return type. ## What file formats can I upload for the draft return? The draft return must be a PDF. The uploader accepts one or more PDF files. Binder source documents can be PDFs, images, Excel files, or Word documents, but the draft return specifically must be in PDF format. ## How long does AI Review take? AI Review typically completes in a few minutes. The page updates automatically when the review is ready; no manual refresh is needed. Processing time varies with the number and size of source documents in the binder and the complexity of the return. ## What if the binder is missing documents? The binder preview screen on step 2 of the initiation flow shows missing document flags for the client's configured checklist. You can open the binder in a new tab from that screen to upload missing documents without losing your place in the review flow. Running AI Review with an incomplete binder is allowed, but findings are limited to the evidence that is present. ## What do the severity levels mean? * **Critical** - A finding that should be resolved before the return is filed. * **High** - Strong evidence of a discrepancy between the draft and source documents. * **Medium** - A potential issue that warrants review; the evidence is present but the discrepancy may be explainable. * **Low** - An informational or low-confidence finding that is less likely to require a correction. Severity is assigned by the AI and should be treated as a starting point for the reviewer's judgment, not a final determination. ## Can I run AI Review more than once? Yes. You can start a new run at any time using the "+" button in the AI Review sidebar. Common reasons to re-run: new source documents were added to the binder, or a revised draft return is available. ## What happens to previous results when I re-run? Each run is independent and remains visible in the run history sidebar. The most recent run becomes the active set of leadsheets and the active set of binder flags. Earlier runs remain accessible by selecting them in the sidebar; they are not deleted. ## A finding looks wrong - what should I do? Use the comment thread on the finding card to note the discrepancy and discuss it with colleagues. If the finding is a false positive, document that reasoning in the comment thread. A reviewer can sign off the row to mark it cleared even if the underlying item was determined to be a non-issue. ## How do AI Review results relate to the Leads tab in the binder? AI Review writes its findings directly to the Leads tab as leadsheets, one per form on the return. The Leads tab is the working surface for reviewing and clearing findings; AI Review is what creates them. Any binder user with access to the client can see and work through the leadsheets once the review completes. ## Does AI Review replace the manual review process? No. AI Review is a first-pass assistant that handles the cross-referencing and evidence-gathering steps that would otherwise be manual. The preparer still reviews every finding, exercises professional judgment, and signs off on each item. AI Review accelerates the process; it does not replace the CPA's role. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.filed.com/reference/products/ai-review/frequently-asked-questions.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.