A ChatGPT screenshot gives itself away instantly, so turn on Docognito before you capture it. The logo, the header and the chat bubbles are recognisable at a glance, and Docognito redraws the page as a document editor before you press capture, so the image already matches rather than needing edits after the fact.
What a raw screenshot gives away
A screenshot has no context and no explanation attached to it by default. The person receiving it over Slack or email sees exactly the pixels you sent, and a ChatGPT page hands over a lot in those pixels: the green or black logo mark, the rounded chat bubbles, the header bar with the model name, and the layout that anyone who has used the tool recognises within a second.
None of that is a problem in every context. It becomes one the moment the recipient is a client, a manager, or a channel where AI use has not been discussed openly, because the screenshot answers a question nobody asked. The content of the answer might be exactly what the message needed to say, useful research, a drafted paragraph, a checked calculation, and the wrapper around it is the part that changes how the message reads.
There is also the question of what happens after you send it. A message sent to one person can be forwarded, screenshotted again, or pasted into a wider channel without your knowledge, and each of those copies carries the same logo and layout with it. Fixing the look once, before the first send, means every later copy inherits the same document appearance rather than needing to be caught and re edited each time it resurfaces.
Turn on the disguise before you capture
The order matters here. Capturing first and editing afterwards means cropping a logo out of a still image, which is slower and never quite matches the rest of the page. Turning the disguise on first means the screenshot is already correct the moment you take it.
Switch the page to its document look
Press the shortcut, or use the toolbar toggle, before you open your screenshot tool. The page redraws as a document editor in under a second.
Give the page a moment to settle
Wait for the redraw to finish before capturing. A screenshot taken mid transition can show a half changed page rather than the finished layout.
Capture the page area only
Use a window or region capture rather than a full screen capture, so the browser chrome around the page stays out of the image entirely.
Crop out the address bar too
A full window screenshot still carries the address bar inside it
Docognito changes the page, the tab title and the favicon. It does not, and cannot, change the address bar, so the address bar, which still reads claude.ai or chatgpt.com. A screenshot of the entire browser window includes that bar as part of the image, and no amount of page disguise removes it once the capture is taken.
A region capture that starts just below the address bar, or a window capture tool with the option to exclude browser chrome, solves this in one step. If your screenshot tool only offers a full screen capture, crop the image afterwards before you send it, since that single strip of pixels is the one part of the picture the disguise was never able to reach.
Screenshot tools and what they capture
A full screen capture takes everything on the display, taskbar and all. A window capture takes the browser window, address bar included. A region capture, drawn by hand around the part of the screen you actually want, is the only mode that reliably leaves both the address bar and the rest of the browser chrome out of the image. On most systems this is a built in shortcut, not a separate download, which makes it the simplest option to reach for by habit rather than something to remember to install.
Some screenshot tools also offer a delayed capture, counting down a few seconds before taking the picture. That delay is useful here for a second reason beyond giving you time to arrange the window: it also gives the page time to finish its own redraw, so the countdown and the settling time from the earlier step can happen together instead of one after the other.
Sending it afterwards, not sharing it live
This is a different situation from screen sharing on a call. The image sits in a message thread, gets forwarded, and may be opened again months later by somebody who was never in the original conversation. A screenshot lives longer than a glance across a desk, which is exactly why the page should already look right before the shutter closes rather than depending on the recipient never zooming in.
Get the sequence right before you press capture
Toggle the disguise, wait a second, then capture the page area only, leaving the address bar out of frame or cropped afterwards. Add Docognito once and the first two steps take a single keypress every time you need them again.