I was somewhat worried about that until I saw this: twitter.com/nickjshearer/status/1409902…?... I think programming is one of the many domains (including driving) that will never be totally solved by AI unless/until it's full AGI. The long tail of contextual understanding and messy edge-cases is intractable otherwise. Will that happen one day? Maybe. Will some kinds of labor get fully automated before then? Probably. But I think the overall time-scale is longer than it seems.

twitter.com/eevee/status/14100491950676… """ "but eevee, humans also learn by reading open source code, so isn't that the same thing" - no - humans are capable of abstract understanding and have a breadth of other knowledge to draw from - statistical models do not - you have fallen for marketing

Nope, they explicitly note that the GPL showed up 700k times in the training data: twitter.com/eevee/status/14100678602992…

Great article. Necessary now more than ever. Those of us in the FOSS community now need to also move away from GitHub. They are taking our code and putting it into their ML-based copilot product to allow proprietary developers to use FOSS code without having to conform to a FOSS license. This practice makes GitHub is hostile to the FOSS community. Decentralization is key to preserving our software rights. twitter.com/eevee/status/14100373098487…

The 0.1% figure comes from OpenAI: twitter.com/eevee/status/14100373098487… (person disagrees w/ me, the image is what's relevant). I talk about Lodash one-liners because it makes it pretty obvious that Copilot is not just copy-pasting code (which would be copyright infringement). It's quite unlikely (even if we consider all variables to have the same name), that Copilot is copy-pasting an exact copy of some other snippet, given that the snippet is very specific to my problems. (I'm not asking it to write a generic math function, I'm asking it to use lodash on data structures in my codebase to accomplish a very specific outcome). By quite unlikely I mean (1 / (2 ^ 32)), if we consider the one-liner to be composed of 32 different AST nodes. > Is a Markov chain "Creative"? That's essentially what you're describing here. Have you read the paper that (eventually) inspired Copilot? "Attention is All You Need". It's not a Markov chain. There's many, many, many different steps / layers. I think if you know & understand the fundamental building block that is responsible for it working (the "transformer"), then a lot of the worries around plagiarism go away. Also. What is coding if not a search problem over a very large space? I'm searching for the next N lines to write over the space of all possible lines. To guide my search I use things like my prior knowledge. In my head, this knowledge is encoded using neurons. In Copilot, the knowledge is encoded in parameters. Abstract concepts in my head are encoded in neurons. In Copilot's "head", abstract concepts are encoded using "embedding vectors" of 512 bytes (maybe more / less, not sure). Yeah, maybe I use more than 512 bytes to encode a concept, but still, I don't see a huge difference. If I'm not plagiarizing, then I can't imagine Copilot is (except for in the 0.1% of cases where it's over-fitting)

I've not seen Copilot in action yet, I was under the impression it doesn't use code directly. In any case my original question was answered by the tweeter in a later tweet I missed twitter.com/eevee/status/14100491950676… I get where they're coming from but they are kinda just handwaving it back the other way with the "u fell for marketing idiot" vibe. I wish someone smarter than me could simplify the legal ramifications around this but we'll probably have to wait till it kills someone (or at least costs someone a bunch of money) to get any actual laws set up.

Your question had already been preempted in the OP. Specifically: > "but eevee, humans also learn by reading open source code, so isn't that the same thing" > - no > - humans are capable of abstract understanding and have a breadth of other knowledge to draw from > - statistical models do not > - you have fallen for marketing -- twitter.com/eevee/status/14100491950676…

1. Someone expressed their concern that fragments of GPL code would end up in projects with non-GPL licenses and called it open source laundering. They were also concerned about the interpretation of derivative work. Could copilot's output be considered a derivative work? twitter.com/eevee/status/14100373098487… 2. Someone posted how copilot reproduced the famous fast inverse square root function verbatim, including comments. news.ycombinator.com/item?id=27710287