[LanceJones]

OpenAI could be profitable (easily) if it stopped training new models. Whether they will make that choice or not, who knows.

[aatd86]

That would be short-termist though. So, quite unlikely. In my usage (code) they are still better than everything else I have tried. Point being that I am looking predominantly for the one llm that gives me the best code output. If they risk losing that advantage for immediate profit, guess I will cancel like I did for claude... (I still got a gemini subscription, for some reason it has a good UI, fast for common non technical requests).

Seems to have been my pattern of behavior with all these tools.

[johnnyanmac]

>If they risk losing that advantage for immediate profit, guess I will cancel

We call that "when the bubble pops". Can't wait.

[pier25]

Google and Anthropic probably won't stop training though.