Every open source license grants permission for AI training, and GitHub copilot by default rejects completions that exactly match code from its training. You can’t pretend to be pro-open source or pro-free software but at the same time be upset that people are using licensed software within its license terms.
Not all projects on GitHub use the same open source license. I don’t have a problem with scraping on projects that allow it. I have a problem with scraping on the ones that don’t.
Only if you can reasonably argue that the output is the input (even with exact matches over a certain size being auto-rejected), and that it is enough to qualify as a copyrightable work. I’d argue line completions can never be enough to be copyrightable, and even a short function barely meets the bar unless it is considered creative in some way.
Every open source license grants permission for AI training, and GitHub copilot by default rejects completions that exactly match code from its training. You can’t pretend to be pro-open source or pro-free software but at the same time be upset that people are using licensed software within its license terms.
Not all projects on GitHub use the same open source license. I don’t have a problem with scraping on projects that allow it. I have a problem with scraping on the ones that don’t.
If a license forbids LLM training, it is by definition not open source.
Code being visible for anyone to see is open source. The license for that code has nothing to do with it. You’re thinking of FOSS.
If you use agplv3 for training your LLC, shouldn’t the code you spit out also be agplv3?
Only if you can reasonably argue that the output is the input (even with exact matches over a certain size being auto-rejected), and that it is enough to qualify as a copyrightable work. I’d argue line completions can never be enough to be copyrightable, and even a short function barely meets the bar unless it is considered creative in some way.