We provided mental health support to about 4k people – using GPT-3 (twitter.com)

Not OP, this is highly unethical.

Is this allowed per the ChatGPT terms of service? What happens if someone discovers they received support from a computer?

Not to sound like a Ted Kazinsky but I’m pretty sure large language model technology will be unsuccessful at filling the emotional, physical, chemical, or spiritual void a human being with mental health issues is feeling.

This is why ethical research requires human subject approval.

This is honestly just what I would expect from a former AirBnB exec. I'm glad he's getting dragged by so many people; it's very obvious that he sees mental health service provision as little more than a large market to be cornered.