/ OCR errors / legal names / supplier identity
AI OCR Errors in Legal Names Still Break Supplier Files
Why legal names, bank beneficiaries, and registration numbers deserve manual confirmation despite better OCR tools.
Document intelligence keeps improving, yet AI risk guidance still treats measurement, mapping, and management as ongoing work. That headline matters only after it reaches a buyer's desk, a finance queue, or a risk file. Supplier files show why: one bad OCR read can connect the wrong entity to a bank account, certificate, or shipment. The immediate job is not to repeat the news. The job is to decide which supplier record now deserves a harder look, which payment should wait, and which piece of evidence can survive a later question from a manager, broker, auditor, or platform team.
The bad habit is to trust extracted text because the document image appears clear on screen. The better habit starts with one narrow question: what would have to be true before this supplier decision can move forward? That keeps the review from turning into theatre. A team can read a dozen warnings and still release a weak payment if the beneficiary line, legal entity, and source record stay unchecked. A team can also freeze a good order for no reason if every alert becomes a crisis.
Mark legal names, registration numbers, certificate numbers, and bank beneficiaries as high-consequence fields. The reviewer should write that first move into the case file before opening extra tabs. A short entry such as "bank beneficiary changed after invoice approval" or "forced-labor tracing incomplete for named material" is enough. It tells the next person what changed, why the file reopened, and which evidence should settle the point. Vague labels such as high risk or urgent supplier issue do not help anyone.
The useful fields are concrete: raw image, extracted value, reviewer-corrected value, source document, field type, downstream use, and final approved value. These fields do more than fill a checklist. They stop a model, a supplier, or an internal reviewer from hiding behind a general conclusion. If the answer depends on an invoice, name the invoice. If the answer depends on a registration record, show the searched name and date. If the answer depends on a call, record who called, which route was used, and what still needs written proof.
AI can prefill the fields and compare them across documents faster than a person can. That is useful work, but the model should not become the person who clears the case. The output should show the source, the extracted value, the conflict, and the reason the conflict matters. A confidence score without source evidence gives the file a polished look and weak support. For supplier verification, polish is a poor substitute for a traceable record.
A reviewer should confirm every high-consequence field against the image before the value enters a supplier master or payment file. This line should be visible in the workflow, not buried in a policy. The reviewer can accept a field, correct it, reject a match, ask for a second document, or hold the case. Each action should leave a small mark in the file. When a later dispute appears, the team should be able to show what the system found and what a person decided.
Before closing the review, the case owner should test the conclusion against the first move: mark legal names, registration numbers, certificate numbers, and bank beneficiaries as high-consequence fields. If the conclusion cannot point back to that action, the file has drifted. A tidy summary, a long email chain, or a vendor dashboard can make drift hard to notice. The safer closeout names the open field, the accepted field, and the decision that remains blocked until better evidence arrives.
Ask for a clearer scan or original-language document when the field controls identity, payment, or compliance status. A supplier who has the record can usually answer a precise request. A supplier who answers around the request gives the buyer useful information too. The file should keep both outcomes. Silence, delay, a replacement PDF, or a new contact from another domain may matter more than the document itself. Those details often explain why a clean-looking record still needs review.
A simple note says: OCR misread final digit in registration number; corrected from license image; prior match result rerun. This kind of note sounds ordinary, which is the point. It gives finance, sourcing, or compliance a decision they can use without retelling the whole case. It also prevents the review from drifting into reputation language. The file does not need to call the supplier good or bad. It needs to state which evidence supports the next action and where the limit sits.
The correction should stay in the file. Deleting the error hides a useful model-quality signal. The operating rule is simple enough to repeat on a busy day: let AI organize the file, but keep proof and judgment separate. The news cycle will keep changing. The case file should still answer the same questions: who is the legal party, what changed, which source proves it, who reviewed it, and what decision is allowed. Better OCR reduces typing, but it does not remove responsibility for the fields that move money or risk.