alexbecker
Joined 12 years ago
Karma
1,107
Submissions
365
Member since
Aug 26
https://alexcbecker.net
View on Hacker NewsRecent submissions
Comment · 11 months ago
I'm working on _prompt injection_, the problem where LLMs can't reliably distinguish between the user's instructions and untrusted content like web search results.
Just published a blog post a few minutes ago: https://alexcbecker.net/blog/prompt-injection-benchmark.html
Comment · 1 year ago
Generally the threat model is that a trusted user is trying to get untrusted data into the system. E.g. you have an email monitor that reads your emails and takes certain actions for you, but that means it's exposed to all your emails which may trick the bot into doing things like forwarding password resets to a hacker.
Comment · 1 year ago
There have been attempts like https://arxiv.org/pdf/2410.09102 to do this kind of color-coding but none of them work in a multi-turn context since as you note you can't trust the previous turn's output
Comment · 1 year ago
The problem is there is no real way to separate "data" and "instructions" in LLMs like there is for SQL
Comment · 1 year ago
I doubt Comet was using any protections beyond some tuned instructions, but one thing I learned at USENIX Security a couple weeks ago is that nobody has any idea how to deal with prompt injection in a multi-turn/agentic setting.
Comment · 1 year ago
Lately I've been trying to detect/mitigate prompt injection attacks. Wrote a blog post about why it's hard: https://alexcbecker.net/blog/prompt-injection.html
Comment · 1 year ago
After reading Judith Butler for a class in college, reading "Professor of Parody" was such a breath of fresh air. Nussbaum is a clear thinker who doesn't take BS kindly.
Comment · 7 years ago
The argument as I understand it is that group 2 mostly does not care about being protected from their government or even agrees with the government's actions, and they get to benefit from continued access to Apple's superior products. I'm not endorsing this argument, but it's not prima facie crazy.
Comment · 7 years ago
I run a similar, maybe even more boring stack for my less-than-one-person company [PyDist](https://pydist.com):
- PostgreSQL database
- Nginx proxy in front of Django apps for UI and API servers (I use gunicorn instead of uWSGI though)
- Cron jobs which invoke django-admin commands to keep the PyPI mirror in sync
Perhaps the only place I'm any fancier than OP is that my deploy script is in Python, not shell, since any time I try to write a shell script with even slightly nontrivial logic it falls over and catches fire :)