Free one-week check

Run Safe-Desk for a week and see what your staff are giving to AI.

Updated September 24, 2026 · SOPHIA XT LLC

How the week works

1. Install as normal. Sign up, add your staff, and put the Safe-Desk extension in each computer's Chrome or Edge, or the Windows app for the ChatGPT, Claude and Copilot desktop apps. The setup docs give the order for an office.
2. For 7 days, Safe-Desk mostly reports. When a message holds a client detail, such as an account number, an address or a client's name, it goes to the AI as your staff wrote it, and Safe-Desk records what kind it was, who sent it and which AI site it went to. Social Security, taxpayer and card numbers stay hidden all week, because those are the ones that do real harm if they leak, and they are still counted in the report. Passwords, PINs, API keys and text marked privileged are still stopped. File uploads to AI sites follow your upload setting, which stops them unless you change it.
3. You watch the catches arrive. Each one appears in your log, and on your phone if you turn alerts on. You can open the report at any time to see the week so far.
4. On day 7 your report is ready. It shows how many messages held client details, what kinds of data they were, who sent them, which AI sites they went to and on which days, and how many of your people were covered compared with how many you invited. The single biggest finding is written at the top in one sentence. It never shows the values themselves, only what kind of data each one was. You can print it or save it as a PDF.
5. Then protection turns on. The report ends with one button, Turn on protection, which you can press at any point in the week. If you don't, protection turns on by itself at the end of day 7 and you get a notice. From then on client details are swapped for placeholders before the AI sees them, under the rules for your industry.

What the check can't see

Safe-Desk only sees AI use on computers where the extension or the Windows app is installed and signed in. A staff member using AI on a phone, on a home computer, or in a browser without the extension is not in the report. The report says how many of your people were covered, so you can tell how complete the picture is. Names, addresses and dates found from context are a best effort; the client list you add is the sure way to catch names.

What happens to the data

The same as everywhere else in Safe-Desk. Messages and AI answers are never stored. The log keeps who, when, where and what kind of data a rule caught, never the value. During the week, messages reach the AI provider as your staff wrote them, which is what happens today without Safe-Desk. The security page lists what is kept and for how long.

What it costs

The check is the first week of the free trial. No card is needed to start. Add one in Account before the 14 days are up to keep Safe-Desk on; if you don't, it stops at the end of the trial and nothing is charged. Safe-Desk costs $19 per staff member a month, with a 14-day free trial. Each seat includes $5 of AI a month inside Safe-Desk, pooled across the firm, and AI beyond that is billed at the model's cost plus 30%. Using your own ChatGPT or Claude through the extension or the connector costs nothing extra. Optional setup: we set it up for $149 for up to 10 staff, $10 for each extra person.

Related

• How to stop staff pasting client data into ChatGPTPolicy, approved accounts, a gate in the browser, and a log the owner reads.
• Is it safe to use ChatGPT with client data?What happens to what you type, business versus consumer plans, and the rules that apply.
• SecurityWhat we keep, what we never keep, and every subprocessor.