It sounds actually very funny to try and break it
VintageGenious
joined 1 year ago
Even though I hate a lot of what openAI is doing. Users must be more informed about llms, additional safeguards will just censor the model and make it worst. Sure they could set up a way to contact people when some kind of things are reported by the user, but we should take care before implementing a parental control that would be equivalent to reading a teen's journal and invading its privacy.
Some companies make AI interviews, prompt breaking it to pass the interview xD