| ▲ | jrflo a day ago | |
Frontier models are smart enough to ignore naive prompt injection attacks like that. I don’t know if that’s actually a concern or not, I had Opus try download a dataset for me and it found a redirect loop designed to catch AI crawlers and identified it. | ||
| ▲ | noodletheworld a day ago | parent | next [-] | |
I can assure you, personally, Opus is not immune to prompt injection attacks. | ||
| ▲ | coeneedell a day ago | parent | prev [-] | |
100% of the time? | ||