GitHub Copilot’s $9B Lawsuit: What’s Behind the AI Code Controversy?

In June 2021 Microsoft and OpenAI announced the launch of GitHub Copilot, an AI programming tool that quickly excited developers with its code‑completion capabilities.

One year later, as Copilot entered commercial pricing of $10 per month (or $100 per year) with free tiers for students and open‑source maintainers, criticism grew over the fact that many users discovered their code appearing in Copilot suggestions without permission.

Writer, designer, programmer, and lawyer Matthew Butterick filed a class‑action lawsuit in the U.S. District Court for the Northern District of California against Microsoft, GitHub, and OpenAI, alleging that Copilot unlawfully uses open‑source code.

Butterick created the website githubcopilotlitigation.com to track the lawsuit, stating that Copilot relies on “unprecedented software piracy” and that AI must be fair and ethical for everyone.

1. Issues with Training Data and Licenses

Most open‑source packages are released under licenses that require attribution and preservation of copyright notices. Users of such software must:

Comply with the obligations set by the license.

Rely on “license exceptions” such as fair use under copyright law.

Microsoft and OpenAI acknowledge that Copilot and its underlying model Codex were trained on public GitHub repositories, but they have not provided the required attribution for the licensed code.

The defendants therefore rely on a “fair use” defense, which GitHub’s former CEO Nat Friedman previously described as permissible for training on publicly available data.

2. Issues with Code Completion and Liability

Microsoft describes Copilot’s output as “code suggestions” and claims no rights over the generated code, while also stating that it does not guarantee correctness, safety, or freedom from intellectual‑property issues.

Consequently, users bear full responsibility for the security and quality of any code accepted from Copilot, and are advised to apply the same testing, IP scanning, and vulnerability tracking as they would for any third‑party code.

Various parties, including the Software Freedom Conservancy (SFC) and Texas A&M professor Tim Davis, have publicly criticized Copilot for violating open‑source licenses and copyright.

The lawsuit alleges violations of GitHub’s Terms of Service, DMCA §1202 (prohibiting removal of copyright management information), California Consumer Privacy Act, and other applicable laws.

The complaint seeks statutory damages of $9 billion, arguing that each of the 1.2 million Copilot users could trigger multiple DMCA violations, amounting to billions in damages.

Butterick indicates that the complaint may be amended to add more defendants and claims, emphasizing that this is the first U.S. class‑action targeting an AI system’s training and output, and that AI developers must be held accountable.

While many developers doubt the lawsuit’s chances of success, they hope it will prompt platforms to improve practices.