(gandalf.lakera.ai)

378 points hubraumhugo | 3 comments | 11 May 23 18:04 UTC | HN request time: 0.632s | source

1. uglygoblin ◴[12 May 23 01:30 UTC] No.35910752[source]▶

This is pretty fun and a clever way to collect wild prompts and analytics about how well their AI safety suite is doing :)

replies(1): >>35912348 #

2. cjonas ◴[12 May 23 05:41 UTC] No.35912348[source]▶

Would be nice if there was an open source version of this, where the data was published for the public to learn from

replies(1): >>35921797 #

3. dh00608000 ◴[12 May 23 20:19 UTC] No.35921797[source]▶

It's something we're actively looking into!

Gandalf – Game to make an LLM reveal a secret password