EPISODE · Jul 15, 2026 · 9 MIN
Overthinking attack slows robots up to 7 times: vulnerability in vision-language models
Ricercatori del Michigan Technological University hanno scoperto una nuova vulnerabilità nei robot che utilizzano modelli vision-language. Un attacco basato su testo strategicamente posizionato può indurre l'AI a un 'overthinking' che rallenta le risposte fino a 7 volte, creando un denial-of-service. La difesa è semplice ma critica per sistemi tempo-sensibili.
Embed this episode
NOW PLAYING
Overthinking attack slows robots up to 7 times: vulnerability in vision-language models
0:00
9:07
1×
No transcript for this episode yet
Similar Episodes
No similar episodes found.
Similar Podcasts
No similar podcasts found.
Frequently Asked Questions
How long is this episode of GoYou Cybersecurity (EN)?
This episode is 9 minutes long.
When was this GoYou Cybersecurity (EN) episode published?
This episode was published on July 15, 2026.
Can I download this GoYou Cybersecurity (EN) episode?
Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!