AI generated content: can you still tell the difference?

[ClusterCanopy]

A few years ago, spotting AI generated content was easy. Images had too many fingers, text was repetitive and bland and videos looked like strange dreams. Now the quality has improved enormously. AI images can look like real photographs, AI voices can mimic real people and AI writing can be hard to distinguish from human writing, especially in short pieces

There are still some telltale signs, at least sometimes. AI writing often uses certain phrases and structures repeatedly, overuses lists and sounds confident without saying much. AI images can have odd details in the background, nonsense text on signs or lighting that does not quite make sense. Video still struggles with physics, like how water moves or how hands interact with objects

Personally, I find text the hardest to judge. I have seen posts I was sure were AI that turned out to be written by people, and the reverse. Detection tools exist, but they are unreliable and can wrongly accuse people, which is a real problem for students. I am less confident than I used to be

The bigger worry is what happens when nobody can tell. Scam calls using cloned voices, fake reviews, fake news photos and AI generated product listings are already common. Some platforms now label AI content, but labelling is inconsistent and easy to strip out

Can you still tell the difference? And what are the signs you look for?