Have a look at this photo.
Now answer this honestly: could you tell this picture was generated by Artificial Intelligence?
After unsuccessfully searching through stock-image sites for a photo I needed to accompany a LinkedIn post, I asked Google's Gemini AI (formerly known as "Bard") to create "a photo of a friendly retail shopping experience, with a store employee being friendly to a customer."
This photo was one of four that came back, but I could have used any one of them.
In all cases, the faces weren't distorted.
All hands shown had the correct number of fingers, a problem that has plagued most AI image-generation tools until very recently.
Sure, if you look carefully at this particular image, you may notice the top of the hanger is a little bit too long and jabbing into the woman's neck, or that the top of the hanger isn't curved to accommodate a clothing rod.
You may notice that both women's eyes look a little "off"; as one person remarked when I asked the "real or not" question on LinkedIn, "It's always the eyes. Can't fake a soul."
But this photo is exponentially better than anything an AI tool could have generated just six months ago.
It was better than any of the stock images of real people I could find.
It was free to generate and use in whatever way I needed.
And I'm not sure I would have noticed it wasn't a real image if I didn't prompt its creation.
It's increasingly difficult to tell if the images you see on the internet are authentic or artificial.
That's exciting and terrifying at the same time.
If you haven't already started to experiment with AI tools, there's no time like today.