ZeroHour
DashboardNewsCVEsExploitsIncidentsAIBriefAnalyticsEntitiesWatchlistIOCsAsk the feedSourcesFeeds & APIKiosk mode
Schneier on Securitypublished ()ingested Bruce Schneier

New Image/Video Prompt Injection Attacks

mediumAI safety & securityimportance 30
Read the original
Tagsai-safety
Full article93 words · extracted from schneier.com · click to collapse

HomeBlog

Simon Willison has been playing with the video processing capabilities of the new Gemini Pro 1.5 model from Google, and it’s really impressive.

Which means a lot of scary new video prompt injection attacks. And remember, given the current state of technology, prompt injection attacks are impossible to prevent in general.

Tags: AI, Google, LLM, video

Posted on February 22, 2024 at 12:08 PM • 26 Comments

Comments

Atom Feed Subscribe to comments on this entry

Sidebar photo of Bruce Schneier by Joe MacInnis.

Powered by WordPress Hosted by Pressable

Text extracted automatically; images, tables and formatting may be missing. Original: https://www.schneier.com/blog/archives/2024/02/new-image-video-prompt-injection-attacks.html