Maybe Copilot could be turned into a context-aware search engine? That is, invoking it would return a list of examples that it thinks do the same thing as what you're trying to do, based on your work-in-progress code.
Seems like humans are more creative when they can get inspiration from language models.
Ironic if AIs end up being more useful for creativity than for detail-oriented tasks.
Re:Risk Assessment of GitHub Copilot
by jessmartin on Monday July 12, 2021 @09:46PM
(#27816938)
I honestly think this is where things are going. Man-machine partnership on creative tasks. Not only is it more amenable to current models, it’s higher leverage and less likely to be completely automated away.
I wrote a little about this shift to what I think of as “conversational programming”: https://jessmart.in/articles/copilot
You're not going to get it out of a generation of people raised by tablets, needs to come from somewhere.
This would be way more interesting of a product than what copilot actually is.
Yeah: I would love this; I actually often try to use GitHub like this, and it sort of works at times, except GitHub's search features frankly suck.
I would have thought to pivot to an assistive model in the case you mention, and in cars and in copilot.
Have the machine notify you when it thinks you’ve made a mistake.
Re:Risk Assessment of GitHub Copilot
by mumblemumble on Monday July 12, 2021 @04:31PM
(#27814722)
The thing we ran into was, in cases where the human and the machine disagreed, it was basically never the human that was wrong.
So we pivoted the product to being something you would run on full auto, for situations where you didn't need a high level of quality. I'm not sure if that option is available to programmers, though.