2 ms·
I thought I didn't have this yet but turns out it's there, just buried in the drop down model selection menu. I tried it out on an existing chat thread where I
by modeless 2y ago
I thought I didn't have this yet but turns out it's there, just buried in the drop down model selection menu. I tried it out on an existing chat thread where I was writing JavaScript and when I said "use canvas" it started rewriting my JavaScript to use HTML5 <canvas>. I had to tell it "use canvas, not html5 canvas but chatgpt canvas" and then it figured it out.
It seems like this will be perfect for making tiny single page HTML/JS apps.
- xp84 2y agoThat anecdote makes me wish we could make LLMs less compliant … I’m imagining some code wildly inappropriate to shoehorn a <canvas> into and GPT happily complying, instead of responding “are you nuts? How would a <canvas> be useful to this method that validates a certificate?” Obviously it’s harmless here, but I can picture “evil genie” results when someone asks a question to an LLM entrusted with important privileges. Like if you said “Open ‘uranium containment’” meaning a file with that name, and the LLM is like “don’t want to argue with humans!” opens up the uranium containment doors An extreme and stupid example, obviously, but you get the idea. If I am to trust an LLM it ought to be able to say “That sounds like a stupid idea. Surely we have a misunderstanding.”