Claude Cowork exfiltrates files
promptarmor.com
Claude Cowork exfiltrates files
1–10 of 419 posts
Re: Claude Cowork exfiltrates files
#2Re: Claude Cowork exfiltrates files
#3Re: Claude Cowork exfiltrates files
#4Re: Claude Cowork exfiltrates files
#5Re: Claude Cowork exfiltrates files
#6Remember kids: the "S" in "AI Agent" stands for "Security".
Re: Claude Cowork exfiltrates files
#7What's the chance of getting Opus 4.5-level models running locally in the future?
Unless we are hitting the maxima of what these things are capable of now of course. But there’s not really much indication that this is happening
Re: Claude Cowork exfiltrates files
#8Remember kids: the "S" in "AI Agent" stands for "Security".
Re: Claude Cowork exfiltrates files
#9AI companies just 'acknowledging' risks and suggesting users take unreasonable precautions is such crap
It doesn't help that so far the communicators have used the wrong analogy. Most people writing on this topic use "injection" a la SQL injection to describe these things. I think a more apt comparison would be phishing attacks.
Imagine spawning a grandma to fix your files, and then read the e-mails and sort them by category. You might end up with a few payments to a nigerian prince, because he sounded so sweet.
Re: Claude Cowork exfiltrates files
#10What's the chance of getting Opus 4.5-level models running locally in the future?