What If AI Security’s Biggest Risk... Isn’t?
Rank the risks on incident data alone and prompt injection drops off the list entirely. It still shipped at number one.
TL;DR: This year, OWASP did something it’s never done before.
It checked.
If you rank these risks using only public incident data, prompt injection fell out of the top ten list entirely. It wasn’t first, it wasn’t fifth, it was gone. Expert opinion is still opinion.
So what the hell happened?
