THREAT OPS › Threat News › AI threats in the wild: The current state of prompt injections on the web
AI threats in the wild: The current state of prompt injections on the web
<span class="byline-author">Posted by Thomas Brunner, Yu-Han Liu, Moni Pande</span><div><br /></div><div><span id="docs-internal-guid-49e83394-7fff-3df2-a974-c7291d12beb8"><p dir="ltr" style="line-height: 1.38; margin-bottom: 10pt; margin-top: 0pt;"><span face="Roboto, sans-serif" style="font-size: 12pt; font-variant: normal; vertical-align: baseline; white-space: pre-wrap;">At Google, our Threat
MITRE ATT&CK techniques
- VulnerabilitiesT1588.006
- Social MediaT1593.001
- Generative AIAML.T0016.002
Indicators of compromise
- https://greshake.github.io/url
- https://commoncrawl.org/url
- https://swisskyrepo.github.io/PayloadsAllTheThings/Prompt%20Injection/#toolsurl
- https://research.google/pubs/an-introduction-to-googles-approach-for-secure-ai-agents/url
- https://deepmind.google/discover/blog/advancing-geminis-security-safeguards/url
- blogger.googleusercontent.comdomain