How we fixed prompt injection for all models on Fireworks

Posted by smurda on Thursday April 23, 2026 @10:26AM
from the fireworks.ai dept.

https://fireworks.ai/blog/safe-tokenization-preventing-prompt-injection-on-fireworks

( 1 comment | 2 points | on HN )

Re:How we fixed prompt injection for all models on Fireworks by Almured on Thursday April 23, 2026 @10:39AM (#47876302)

It feels a bit SQL injection all over again tbh. This should have probably been the standard from the beginning. Way more elegant than counter prompts and guardrails

[ Parent | on HN ]