Github Piracy

[Tamela]

GitHub Copilot sounds like a fascinating tool on paper. Launched in preview by Microsoft over a year ago, it's basically an AI-powered extension that uses the code present in all public repositories on GitHub and then write code on your behalf based on simple prompts. For example, you could just write "Take an average of the numbers in this list" and GitHub Copilot will autonomously write the code to do this based on its training on GitHub codebases.

Microsoft announced GitHub Copilot's general availability a few months ago at $10/month, but pretty much since its inception, the product has been criticized for its legal status in the sense that it uses the code written by other people without their permission and that Microsoft profits off of these practices. Today, a class-action lawsuit has been filed against the AI-powered pair programming tool.

The lawsuit has been initiated by Matthew Butterick, who is a programmer, author, and lawyer. He is being represented by the Joseph Saveri Law Firm from California. Together, they are claiming that Microsoft is engaging in open-source software piracy by using billions of lines of code written by millions of programmers under various licenses including MIT, GPL, and Apache. The defendants named in the lawsuit are GitHub, Microsoft, and its technology partner OpenAI.

Butterick claims that Microsoft has violated a number of requirements for using this code including attribution, GitHub's own policies, the California Consumer Privacy Act, and DMCA 1202 - which "forbids the removal of copyright-management information". Butterick goes on to say that:

I am grateful to the programmers and users who came forward to bring this case to fruition and ensure that corporations like Microsoft, GitHub, and OpenAI cannot unfairly profit from the work of open-source creators. This case represents the first major step in the battle against intellectual-property violations in the tech industry arising from artificial-intelligence systems. In this case, the work of open-source programmers is being exploited. But this will not be the last community of creators who are affected by AI systems. Our firm is committed to standing up for these creators and ensuring that companies developing AI products are held accountable under the law.

This paper presents a revealed preference approach to evaluate the welfare of the widely adopted software takedown policy that secures copyright in cyberspace. I build an analytical model to distinguish two hypotheses-the piracy hypothesis versus the fair use hypothesis. The piracy hypothesis predicts that copying is an unproductive investment to steal content; it reduces original contributions and copyright owners' revenue. The fair use hypothesis predicts that copying is a cost-saving method to expand and transform content; it invites original contributions and has little effect on copyright owners' revenue. The takedown policy is only socially beneficial in removing the piracy type of reproduction, but it is welfare-reducing in removing the fair use type of reproduction. An event study design using data from GitHub provides evidence that reproduction on this platform is most likely driven by the motive for fair use, and enforcing the takedown policy is thus inefficient. I conclude by discussing possible optimal policies.

Contrary to some belief, this mod does not contain any anti-piracy checks of any kind. Pirated copies just do not receive support. If it works then, it works. If not, additional support is not going to be added.

Create an reframework/autorun folder in your game directory. This is automatically created when REFramework loads. REFramework will automatically load whatever *.lua scripts are in here during initialization.

When a Lua error occurs here, the reason will be written to a debug log. DebugView is required to view these. In newer nightly builds, the errors can be viewed directly within the ScriptRunner window.

The plugins can perform much of what Lua can, with much more freedom. They have access to much of the important SDK functionality of REFramework, as well as useful callbacks for rendering/input/game code.

RE Engine's IL2CPP implementation is not the same as Unity's. While RE Engine and Unity have many similarities, they are not the same, and no existing tooling for Unity or IL2CPP will work on the RE Engine.

C# scripting maybe a possibility in the future for more natural interaction with the engine, but is not currently being looked at. REFramework is open source, so any developer wishing to do that can try.

The issue of generative AI and the ownership rights of content used in the training data is a new legal frontier. In fact, this issue may pose the biggest threat faced by the industry. The technology is working well, and adoption of the applications based on generative AI is growing quickly. That means the technical risk and demand risk for these solutions have been significantly mitigated. The legal risk is unknown. Precedents set by the outcome of this class action lawsuit could have far-reaching implications.

You have to wonder how Microsoft would react if code from Microsoft Excel suddenly started showing up in Copilot for developers creating the next spreadsheet application. I suspect this would be addressed swiftly.

A group calling itself GitHub Copilot Litigation has filed a class action lawsuit against Microsoft and OpenAI for what boils down to software piracy. Copilot is a product of Microsoft\u2019s GitHub division that helps software developers write code faster by providing or suggesting code using the technology behind OpenAI\u2019s GPT-3. This is the problem, according to Matthew Butterick, who has organized the effort against Microsoft and OpenAI.

The suit was filed in U.S. federal court in San Francisco to challenge the \u201Clegality of GitHub Copilot (and a related product, OpenAI Codex, which powers Copilot).\u201D The plaintiffs allege in their filing:

Copilot ignores, violates, and removes the Licenses offered by thousands\u2014possibly millions\u2014of software developers, thereby accomplishing software piracy on an unprecedented scale. Copilot outputs text derived from Plaintiffs\u2019 and the Class\u2019s Licensed Materials without adhering to the applicable License Terms and applicable laws.

Butterick is co-counsel for the class action lawsuit and lists his background as a \u201Cwriter, designer, programmer, and lawyer.\u201D He created three different types of font and a programming language for web publishing. Butterick also has degrees from Harvard and UCLA. In an open letter on gihubcopilotlitigation.com he summarizes the effort saying:

By train\u00ADing their AI sys\u00ADtems on pub\u00ADlic GitHub repos\u00ADi\u00ADto\u00ADries (though based on their pub\u00ADlic state\u00ADments, pos\u00ADsi\u00ADbly much more) we con\u00ADtend that the defen\u00ADdants have vio\u00ADlated the legal rights of a vast num\u00ADber of cre\u00ADators who posted code or other work under cer\u00ADtain open-source licenses on GitHub. Which licenses? A set of 11 pop\u00ADu\u00ADlar open-source licenses that all require attri\u00ADbu\u00ADtion of the author\u2019s name and copy\u00ADright, includ\u00ADing the MIT license, the GPL, and the Apache license.

Butterick told The Verge\u2019s James Vincent, \u201CThis is the first step in what will be a long jour\u00ADney. As far as we know, this is the first class-action case in the US chal\u00ADleng\u00ADing the train\u00ADing and out\u00ADput of AI sys\u00ADtems. It will not be the last. AI sys\u00ADtems are not exempt from the law. Those who cre\u00ADate and oper\u00ADate these sys\u00ADtems must remain account\u00ADable.\u201D

When GPT-3 first emerged, there was concern that the text generator was simply plagiarizing written work from the web. It turns out that this has largely not been the case. Some of that may be the result of a simple fact. You can express the same idea in many different ways using written language. Software code doesn\u2019t offer you the same latitude.

There are certainly multiple ways to formulate code for a variety of tasks, but it tends to be closer to a deterministic exercise as computers require instruction sets to be in a specific format. Writing a poem, essay, or email doesn\u2019t require the same adherence to a narrow set of rules.

The same is likely true for artwork, though there is also controversy about the text-to-image generators that are too closely copying the styles of living artists. \u201CPeople are pretending to be me. I'm very concerned about it; it seems unethical,\u201D said artist Greg Rutkowski in an interview with Business Insider. At the time of the article, there were 93,000 images that included Rutkowski\u2019s name in the prompt. Granted, this is replicating a style as opposed to providing an exact copy.

It is not surprising that OpenAI Codex offers exact replicas of code blocks. If there is a \u201Cstandard\u201D method for writing code in a programming language to execute a function, OpenAI\u2019s training data is going to see that code in the same format over and over again. GPT-3 will also want to provide that code in a format that is highly likely to be correct as opposed to considering variations. The training of the system and the rules associated with the output will bias and text-to-code generator toward an exact replica, whereas this can more easily be avoided with prose and art.

The fact that Microsoft, the long-standing champion of the proprietary software model, is at the center of this controversy as a defendant is noteworthy. Microsoft has, in recent years, heavily promoted the use of open source software. The GitHub acquisition and interest in hosting open source software and services in Microsoft\u2019s Azure cloud have changed the company\u2019s traditional bias toward licensed software.

c80f0f1006