Apple has finally brought Visual Intelligence to the Mac with macOS Golden Gate, and it is a boon when it works. Here's how to get started.
I admit I have sometimes taken a photo of my Mac's screen and used Visual Intelligence on my iPhone to find out what I'm looking at. But as of macOS Golden Gate, I no longer need to do that because the Mac has Visual Intelligence built in.
Apple's Sebastien Marineau-Mes, vice president of Intelligent System Experience Engineering, announced this during the WWDC 2026 keynote. But frustratingly, all he then said was that you could use it with "a dedicated keyboard shortcut."
He didn't say what that shortcut was, but it's Shift-Command-6.
That's an extension of what might be a very familiar series of keyboard shortcuts. If you've been using the Mac for long enough, you may well know that Shift-Command-3 takes a shot of your whole screen, and Shift-Command-4 takes one of a segment you choose.
Then with macOS Mojave in 2018, Apple added Shift-Command-5. This is one keystroke that presents a toolbar with all of the options for screengrabs and also screen recordings.
It's still all of the options, as Visual Intelligence has been added to it. So you can use Shift-Command-6 directly, or select from the toolbar:
- Press Command-Shift-5
- Click on the Visual Intelligence icon
- Then click on any Mac window
- Choose from the options that appear
The real benefit of using Command-Shift-6 is that you have finer control. Instead of capturing a whole window, you can drag your cursor over a section.
Apple has added Visual Intelligence to the Mac and included it in the screenshot toolbar.
Whichever way you do it, when you use Visual Intelligence on the Mac, various options appear.
These options appear as two or more small popouts somewhere around the image, and the first is always "Ask Siri." Click there and you can type your prompt.
It's not very quick, but you do then get an answer shown in the new Siri AI dialog box. Note that so far in AppleInsider testing, the results you get are as likely to be wrong as with any AI chatbot.
The other option that always appears is "Image Search." Click on that and you're taken to a Google search, although at present the feature isn't working.
Results will vary
What is working, at least sometimes, is Visual Intelligence's ability to recognize food. Much of the time if the image you pick is of a meal, you will also get a button labelled "Look Up Nutrition."
Clicking that button will typically identify the meal, describe it, and summarize its nutritional value from Very Low to Very High. It will also list:
- Protein
- Grains
- Fat Type
- Sodium
- and how processed it is
The trouble is that there's no way to say that what you are looking at is food, and Visual Intelligence often does not recognize it as such.
Actually, it's a bacon naan and Visual Intelligence got it right before. But as long as it tells me I'm fine to eat it, I'll live with the error.
Then, too, you should use Visual Intelligence twice on a food photo if you use it at all. You can get totally different results from exactly the same shot.
Plus even when it does offer the "Look Up Nutrition" option, there's no way to copy that information. There is no copy button, and you can't select the text to copy it, either.
So Visual Intelligence on the Mac is limited enough to be frustrating, but it's good that it's there. This is just the first beta, too, so hopefully it will improve.
