greshake/llm-security
View on GitHub ↗ ★ 2,152 Jupyter Notebook
New ways of breaking app-integrated LLMs
8 mentions on Hacker News by 3 people: 3 stories, 5 comments · first 2023-02-26 · latest 2023-05-22
20082026
-
…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:…
comment by greshake · 2023-05-22 · in Compromising LLM-integrated applications with indirect prompt injection -
▲ 238
Show HN: ChatGPT Plugins are a security nightmarestory by going_ham · 2023-03-25 · 41 comments
-
▲ 4
Show HN: ChatGPT Plugins are a Security Nightmarestory by greshake · 2023-03-25 · 2 comments
-
…use cases for LLMs that process untrusted inputs are unsafe. See https://greshake.github.io/ and https://github.com/greshake/llm-security for more information.
comment by greshake · 2023-02-27 · in Ask HN: API developers, what do you think of LLMs? -
…as described by Greshake[1] seems to be much more realistic to me than I initially thought. [1] https://github.com/greshake/llm-security
comment by rnosov · 2023-02-26 · in Show HN: AI Files – manage and organize your files with AI -
…He argued below that he is not vulnerable to indirect prompt injection attacks ( https://github.com/greshake/llm-security ), but I think he is wrong.
comment by greshake · 2023-02-26 · in Show HN: AI Files – manage and organize your files with AI -
…etc. could all be possible. Look at the paper to find more descriptions of what could go wrong: https://github.com/greshake/llm-security
comment by greshake · 2023-02-26 · in Show HN: AI Files – manage and organize your files with AI -
▲ 17
Show HN: LLMs can be susceptible to a new kind of malwarestory by going_ham · 2023-02-26 · 10 comments