Well... the whole training set is licensed, so you can't really get rid of it. I think that the technology they are using for this is just not ready.

Just retrain the model using properly licensed code? ("just" is doing a ton of heavy lifting, but let's be real, that's not impossibly hard)

There's not many licenses that let you reuse code without including the same headers / licensing blurb. You're in public domain, non-copyleft territory. WTFPL etc.

There is no such thing as properly licensed code because it is a function of the what is legally acceptable for your company and what it intends to do with the work.

Unlicensed code just means “all rights reserved.” You’d need to limit it to permissively licensed code and make sure you comply with their requirements.

That would be a long list of restated license terms and attributions.

Which licenses would it be ok that the training material is licensed under, though? If it produces verbatim enough copies of eg. MIT licensed material, then attribution is required. Similar with many other open source-friendly licenses.

On the other hand, if only permissive licenses that also don't require attribution is used, well, then for a start, the available corpus is much smaller.

The overwhelming majority of code on Github, even code under permissive licenses, require attribution of the original authors.

How would they do that?

Read the LICENSE file in each repo.

What guarantees it’s intact?

It doesn't need to be. If the license isn't positively exactly permissive then you can't use it.

Can you even trust that the License in a random repo is accurate and expresses the actual copyright of all the contained code?

I guess my point is, you can't be positive that even if you're following the license in a repo you forked that the repo owner hasn't already violated someone else's license, and now transitively, so have you.

> Can you even trust that the License in a random repo is accurate and expresses the actual copyright of all the contained code?

In fact, that seems to be exactly the problem shown in the tweet - someone copy-pasted the quake source and slapped a different license on it, and copilot blindly trusted the new license.