This is a great example - it kinda makes sense if you skim read it but butterflies have nothing to do with butter, just like hotdogs have nothing to do with dogs.

I found that trying “some-nonsense-phrase meaning” won’t always trigger the idiom interpretation, but you can often change it to something more saying-like.

I also found that trying in incognito mode had better results, so perhaps it’s also affected by your settings. Maybe it’s regional as well, or based on your search result. And, as AI’s non-deterministic, you can’t expect it to always work.

After being virtually dead, it’s had a lot more development over the last few years(?), with steady progress towards passing the tests and supporting the specs (including reporting spec bugs and vagueness). It’s still a long way from being generally usable.

The focus is on making something that could be embeddable, although there are basic browsers using that embed. The focus seems to be on for use-cases like Electron, which doesn’t need all of the web APIs.

I don’t use it or contribute (yet), but I have their blog in my RSS reader and so keep an eye on it.