[jimmy2020]

There should be a new open-source license specifically designed to restrict the use of code for commercial purposes, whether it involves training machine learning models or not. It could be similar to GPL but tailored specifically for the field of machine learning, ensuring that the code cannot be utilized for ML commercial purposes or trained without limitation.

[layer8]

People aren’t necessarily opposed to commercial use (which many licenses allow), but are upset that license terms like attribution requirements are completely being ignored by Microsoft/Github, turning Copilot into a sort of laundering machine for open-source content.

[jimmy2020]

That's what I am trying to say. Personally, I don't mind if my code is being used for commercial or not. I do mind when Microsoft uses my code without a reference and asks me to pay for copilot just because they decided to put it behind a paywall.

[FrozenSynapse]

Why?! Why restrict the advancement of a field? Most of the times the code uploaded to github came from another source (not original content, very little code is original).

Nobody can stop this anymore. If it's not github that's only taking code from their own platform, there will come others that scrape the internet and use everything, just like MidJourney. Embrace it.

I'm more against Art AI generators, because they produce end results and will fully replace the need to know how to draw. Copilot is just providing simple snippets, it's not thinking for you, so it won't replace the engineers completely, at least for now.

[jimmy2020]

Still, it's good to be an option. If you see this as an advancement you can choose something like MIT if you don't like your code to be trained by a paid copilot then you can choose something like MIT-Human only.

It's a better solution than don't use GitHub at all. And whether we like it or not, a new license should be introduced to address the new data model.