greshake
Joined 3 years ago
Karma
311
Submissions
83
Member since
Dec 5
Recent submissions
Comment · 3 years ago
(depending on whether the search engine context is connected to untrusted inputs or only your curated database ofc)
Comment · 3 years ago
You're missing the more important vector for prompt injection: Indirect injection through the "search engine context". It's not just a matter of blocking bad user questions to fend off reputational harms. See also my work on https://kai-greshake.de/
Comment · 3 years ago
Check out my blog where I show even more up-to-date techniques and the insane ways vulnerable applications are being deployed: https://kai-greshake.de/
Here I go through all of the unsafe products (including military LLMs): https://kai-greshake.de/posts/in-escalating-order-of-stupidi...
Here you can add prompt injections to your resume for free to get your dream job: https://kai-greshake.de/posts/inject-my-pdf/
Comment · 3 years ago
TLDR: With these vulnerabilities, we show the following is possible:
- Remote control of chat LLMs
- Persistent compromise across sessions
- Spread injections to other LLMs
- Compromising LLMs with tiny multi-stage payloads
- Leaking/exfiltrating user data
- Automated Social Engineering
- Targeting code completion engines
There is also a repo: https://github.com/greshake/llm-security and another site demonstrating the vulnerability against Bing as a real-world example: https://greshake.github.io/
These issues are not fixed or patched, and apply to most apps or integrations using LLMs. And there is currently no good way to protect against it.
Comment · 3 years ago
Neither is possible right now.
Comment · 3 years ago
I just published a blog post showing that that is not what is happening. Companies are plugging LLMs into absolutely anything, including defense/threat intelligence/cybersecurity/legal etc. applications: https://kai-greshake.de/posts/in-escalating-order-of-stupidi...
Comment · 3 years ago
I've written about this extensively. My latest article goes into the consequences. How about going from Prompt Injection to airstrike?
https://kai-greshake.de/posts/in-escalating-order-of-stupidi...
Comment · 3 years ago
Also check out a blogpost on the same subject: https://kai-greshake.de/posts/llm-malware/