Here is what checking one value takes in many offices today. The reviewer reads a heat number on a draft, opens the shared drive, searches for the mill test report, waits for the PDF to load, zooms in, finds the number, and switches back. Then they check the next value, and the next draft.

At 9 a.m. on a Monday, a careful person does that every time. At 3 p.m. on a Friday, with the pile still high and a truck waiting, they start to trust the draft. They skim, or they approve the ones that look right. Some people stop using the tool and go back to typing it themselves, because at least then they know where they stand.

1. People skip checks that are slow

We call this the Friday afternoon test. Would a tired person, late on a Friday with the pile still high, still do the review properly in this software? If the answer is no, people will skip the review when it matters most.

In a safe AI setup, the AI does the first pass on the paperwork. A named person then checks the draft, decides, and signs. That rule is only as good as the check, and the most important safety step in the whole system gets done by a tired person late in the week.

2. A wrong number looks just as confident as a right one

The AI drafts a cert packet or a quote from your files. It does not know when it misreads a smudge on a scan. When the order record and the mill cert are clear, the draft matches the record. When something is missing or misread, the AI still writes, and a wrong heat number looks exactly as confident as a right one.

Good software makes the AI flag what it is unsure about. A named person checks each draft against the source before it leaves the building. That check is where a swapped digit on a heat number gets caught. Take it away, or let it turn into a glance and a click, and the rest of the system is a faster way to ship a mistake.

3. Make the right check the easy one

Your shop floor already solves this problem with fixtures. A good fixture only takes the part the right way round, so doing the job right is also the easiest way to do it. Nobody relies on a machinist being extra careful at the end of a double shift. The fixture does part of the remembering.

Review software can work the same way. When the source document is one step away, checking a value takes less effort than skipping it. When the source is five clicks and a second program away, skipping it wins on a Friday, every time.

Take this example. The review screen puts a link to the source file next to every value the AI pulled. The reviewer does not have to click it or open another program. They hover over the link, a large preview of the document appears right there, and they check the value against the source at a glance.

That preview looks like a minor feature. Go back to the Monday-morning routine and count what it removes: the folder search, the second program, the switch back. The check that remains is reading one number against one page, which a tired person will still do on a Friday. Multiply that by every value on every draft, and it is the difference between a review people do and one they learn to skip.

4. Run this test on any tool this week

You can judge review software without knowing anything about AI, and the test works on a tool you are about to buy or one you already have. When a vendor demos a tool, ask to see the review screen. Then, on a Friday afternoon, sit next to the person who will do the reviewing, give them ten real drafts from your own work, and watch for five things. Do not help.

  1. Can they see the source file for a value without leaving the screen? Count the clicks and the programs it takes.
  2. Can they tell which values the AI was unsure about, or does every value look equally confident?
  3. Can they fix a wrong value right where they see it, or do they have to go and edit it somewhere else?
  4. Does the tool record who approved each draft, and can they see the steps the AI took to produce it?
  5. How long did the ten drafts take, compared with doing them by hand? If checking is slower than typing, checking will lose.

A tool that passes all five is one people will keep using properly on a bad week. If a tool fails the first one, people will skim it on a Friday, whatever the demo looked like on a Tuesday morning. Set one house rule before you start. Nobody approves a value they could not see the source for.

This essay comes from Manufacturing's ChatGPT Moment Hasn't Happened Yet: Why AI Fits Your Paperwork and How to Prepare. For how the reviewer's job changes and how to train for it, read Verify, Don't Retype: How Your Team's Job Changes When AI Drafts First.

Questions people ask

Why does software design matter for AI safety?

The safety step is a person checking the AI's draft against the source. If that check means opening other programs and hunting for files, people skim it or skip it late in a busy week. A review screen that puts the source file one hover away keeps the check fast enough that people do it every time.

What should a good screen for reviewing AI work let a person do?

See the source file for a value without leaving the screen, tell which values the AI was unsure about, fix a wrong value in place, and see who approved the draft and what steps the AI took.

Terms in this piece

Friday afternoon test
whether a tired person, late on a Friday with the pile still high, would still review the AI's work properly in a piece of software.