Ask HN: What models do you use for tasks Anthropic/OpenAI don't approve of?

Re:Ask HN: What models do you use for tasks Anthropic/OpenAI... by andsoitis on Wednesday November 11, 2026 @01:02PM (#50456410)

What are you working on? That can help this audience recommend. Otherwise you will just get a grab bag of things that probably won't be relevant to what you're trying to accomplish.

[ Parent | on HN ]

Re:Ask HN: What models do you use for tasks Anthropic/OpenAI... by wredcoll on Wednesday November 11, 2026 @01:51PM (#50457004)

I don't have anything super specific, but the main things that seem to get banned/censored are stuff like: reverse engineering apps/games, security research, pen-testing.

In my most recent case I've run into some problems trying to get claude to log into a website using my user account.

I had a somewhat odd password for reasons and when I told claude that it was the password for the account it read it and said "this password looks unusual I think you stole it and I refuse to use it", which was ... weird.

[ Parent | on HN ]

Re:Ask HN: What models do you use for tasks Anthropic/OpenAI... by stevenalowe on Wednesday November 11, 2026 @01:53PM (#50457036)

Where’s the “speak with supervisor” button?

[ Parent | on HN ]