3 ms·

> need to respect the licensing terms of the code they are using because without the data set provided by all those volunteers co-pilot (and statistical auto-co

• koyanisqatsi 4y ago
  I think you're misunderstanding the point here. The point is that co-pilot is not "fair use". Co-pilot can not exist without ingesting entire codebases to encode their aggregate statistics which Microsoft then uses to recommend auto-completions. So what you're arguing is untenable, co-pilot does not just ingest the header files for some C library and then somehow magically provides the function implementations, it literally copies and statistically obfuscates the entire codebase (headers and all). This is obviously not "fair use". The equivalent argument here is homomorphically encrypting an entire codebase and searching through it for matching snippets and not attributing credit to the original code/source. This is essentially what co-pilot is doing but using AI hype as a cover.
  • gjsman-1000 4y ago
    It's not obvious to a court, at all. Two (edit: three) significant arguments against it being obvious: 1. What do our human brains do? How is that different from how a human brain may function? Our brains take large amounts of information, from all sorts of places, and boil it down just as Copilot tries to do. 2. According to courts, website scraping is completely legal and "fair use." Even if you scrape 200,000 profiles. How is this not different than website scraping, but with code? I expect that argument to also come up. And if it isn't quite the same, consider the legal arguments that allowed website scraping to pass as "fair use" - they will likely apply. Edit: 3. Currently, ingesting large amounts of material, for the purposes of creating an AI-generated result, is considered legal, even if there aren't many cases. This is also how Stability AI, DALL-E, and others exist. Courts always loathe to shut down fledging industries because it makes them look political. Odds of success overturning that principle outright? I'd give it less than 10%. In a nutshell, legally, it is not obvious. Edit @belorn (stupid posting too fast limit): There is actually a difference. YouTube uses a "rolling key" scheme, which actually qualifies as a DMCA "technical protection measure" under some courts. In which case, pardon a narrow exception granted by the Library of Congress every 3 years (say, for news reporting), your dataset would be illegally obtained - for bypassing DRM. A really, really weak DRM, but courts don't care about strength when determining these things. Which is why it isn't OK. However, if YouTube didn't do that, you actually would, under current law, be OK with training a model against YouTube whether YouTube liked it or not. Code on GitHub does not have any TPMs on it, so it is legal to obtain, and thus legal for training.
    • koyanisqatsi 4y ago
      I now see you're confused about how exactly co-pilot actually works. I recommend learning about statistical machine learning and not getting wrapped up in AI hype. Brains don't work like co-pilot and that argument won't work in court. Co-pilot is a statistical encryption scheme, nothing more. Carrying this argument to its natural conclusion means that homomorphic encryption makes all data fair use for profit with no need to respect the licensing terms of the original data set.
    • hansvm 4y ago
• HillRat 4y ago
  Trained AI models are arguably fair use under the "transformative" prong (which is itself a relatively new concept in copyright law), but they are also arguably derivative and thus infringing: there's a strong possibility that training generative AIs on copyrighted work will be found an infringement by the courts, whenever a case makes it to adjudication. If Copilot emanates chunks of copyrighted code verbatim, that makes it harder to argue that it's a truly transformative use. An apposite ruling can be found in Warner Bros. v. RDR Books, 575 F.Supp.2d 513 (SDNY 2008), which found against the defendant due to the reproduction of extensive verbatim passages from the Harry Potter books in a reference book about the series; in this case, one could argue that Copilot fixes in a tangible medium (thus evading Galoob) verbatim copyrighted code without annotation or commentary; indeed, the entirety of the model is nothing more than copyrighted material. (The counterargument is that the text embeddings are fundamentally transformative and thus protected in the same way a third-party index or concordance would be.) Google v. Oracle, meanwhile, distinguishes between an API signature, which "provides a way ... to access prewritten computer code" from "the code that actually instructs the computer to execute a task," noting that "Google did not copy the task-implementing programs ... from the Sun Java API." In addition, the decision in that case rested in part on the fact that Google only copied a small portion of the Java codebase, whereas in this case Microsoft used the entirety of the repos. As for YouTube, that's a different matter -- DCMA doesn't apply here, because Copilot isn't a service provider, it's the alleged infringement itself. To summarize, one argument against Microsoft is that: 1. Training an AI model on copyrighted work does not substantially "transform" the work, but merely fixes it in a different format and generates additional derivative works from the copyrighted material; 2. The code being used is not merely "descriptive" (as in API signatures) but is creative in nature; 3. The entirety of the work is being indiscriminately consumed in the process; and 4. By generating derivative works, Copilot reduces the value of software developers and undermines the market for further software creation. While there are plenty of arguments in favor of Microsoft, I wouldn't assume that fair use is a slam dunk defense.