Document-borne AI worms can self-propagate through Copilot for Word
11–20 of 317 posts
Re: Document-borne AI worms can self-propagate through Copilot for Word
#12Why is it possible to have hidden text in a Word document? Why should the AI have access to that text?
Even if you processed it via a screenshot, image files are processed byte by byte as well and can contain textual metadata.
Re: Document-borne AI worms can self-propagate through Copilot for Word
#13Re: Document-borne AI worms can self-propagate through Copilot for Word
#14It is fun to see how all AI narratives are collapsing.
Re: Document-borne AI worms can self-propagate through Copilot for Word
#15Why is it possible to have hidden text in a Word document? Why should the AI have access to that text?
Re: Document-borne AI worms can self-propagate through Copilot for Word
#16Purged I would have
All things Microsoft from my (controllable) world
Re: Document-borne AI worms can self-propagate through Copilot for Word
#17> Malicious instructions hidden in an externally shared document could make Copilot alter drafted or edited documents in Word and propagate the attack to new documents. Oh no.
And I thought people understood that.
Re: Document-borne AI worms can self-propagate through Copilot for Word
#18> Malicious instructions hidden in an externally shared document could make Copilot alter drafted or edited documents in Word and propagate the attack to new documents. Oh no.
something something lethal trifecta
Re: Document-borne AI worms can self-propagate through Copilot for Word
#19Could this possible be the first AI worm? Or are there any priors to this?