There are certain safeguards and limitations that OpenAI put on ChatGPT to prevent abuse, and to prevent certain ethical or legal problems. Some of these, and the reasoning behind them, were really well explained in the Sam Altman (OpenAI CEO) talks about promises, risks, and fears post video, where the CEO (and some senior staff) talk directly about their fears, worries, hopes, and safeguards. It is very much worth a watch.
There’s a lot of very good reasons why you’d want to prevent an AI from talking romantically (rather than medically) about sex and sexuality. If they implement those kinds of safeguards, while there may be tricks to work around them for a time, or to a very limited degree, those safeguards were deliberate. They monitor for people bypassing them in order to close down those bypasses and tricks. Some of the ways of closing down what they regard as ‘abuse’ is to block IP addresses and access. So be warned that even attempting to find a workaround could lead to loss of access to ChatGPT and even any future products that OpenAI release.
That’s simply not what the tool is for, and they may be pretty determined to prevent people deciding otherwise with their baby.