What a Git History Does and Does Not Prove

Your commit history is a diary, but it isn't a legal deed. You might think that a deep, complex git history proof authorship beyond any doubt, but the Copyright Office sees it differently. They view your repo as a series of technical logs. Those logs show when you changed a file and who was logged into the machine at the time. They don't prove the creative spark behind the code came from you.

The Git Trap

Developers love git. It tracks every character you add or delete. When you look at a commit, you see a clear trail of work. You see the diff, which is the specific block of text that changed between versions. You see a timestamp. You see a username and an email address. You feel secure. You believe this record is the ultimate proof that you wrote the software.

The Copyright Office doesn't agree. In their view, your git history is just metadata. Metadata is data that describes other data. It tells the Office that someone ran a command on a specific day. It doesn't tell them that the code was a human effort. A bot can push to git. A script can generate code and commit it with your credentials. Because of this, the Office treats your commit logs as circumstantial evidence, not as a smoking gun for human authorship.

The Human Spark Requirement

The law currently demands human authorship. This isn't a new rule. It became the bedrock of the system in cases like Thaler v. Perlmutter. In that case, the court ruled that the Copyright Office was right to deny a registration for a work created entirely by an autonomous AI. The machine didn't have a human mind. Without a human mind, there is no author.

Think about Zarya of the Dawn. This was a graphic novel where the human author used an AI tool to generate images. The Office registered the text and the arrangement of the book, but they refused to register the AI-generated images. Why? Because the human didn't have enough control over the final output. The AI made the creative choices.

If you use an LLM or a code-generation tool to write your functions, you're in the same spot. If you paste a prompt into a chatbot and commit the result, you haven't authored the code. You've curated it. The Office wants to know who made the creative choices. If the machine made the choices, you don't own the copyright. Your git history might show you typed the commit message, but it doesn't show the creative process happening inside your brain.

Why Your History Fails the Test

Git history is easy to fake. You can change your git config in seconds. You can set your name to "Linus Torvalds" and your email to a fake address. You can commit code written by an AI and claim it's yours. The Copyright Office knows this. They know that a commit log is a low-effort record.

When you file for registration, you pay a fee. Online registration usually costs about $65. If you're sued for infringement, the stakes get much higher. You could face statutory damages, which are fixed fines set by law. These can range from $750 to $30,000 per work, or up to $150,000 if the infringement is proven to be willful. You want your registration to be bulletproof. If you rely on git history alone, you're building your house on sand.

What Real Authorship Looks Like

If you want to protect your work, you need more than just a repo. You need an audit trail of your creative labor. This means keeping records of your sketches, your pseudocode, and your design documents. It means keeping track of the iterations that show your specific, human-driven design choices.

A real authorship record shows the "why" behind the code. It shows the problems you solved and the specific ways you chose to solve them. When you draft code, try to document the choices that define your style and logic. If you use AI to help you refactor, note what you changed and why you changed it. The Copyright Office is looking for human intent. They want to see that you were the one at the wheel, not the passenger.

Keep in mind that Copyrightable is not a law firm and we don't give legal advice. The rules change fast. You should always confirm the latest guidance at copyright.gov before you file. The current Office position is that they will examine works containing AI content on a case-by-case basis. They will ask you to disclose what was generated by a machine. If you can't show your human input, they will deny your claim.

Beyond the Commit Log

You might be tempted to hide your use of AI tools. Don't. If you lie on your registration application, you risk losing your copyright protection entirely. The Office can cancel a registration if they find out you misled them about authorship.

Your git history is a useful tool for development, but it isn't an authorship contract. It's a technical log. Start documenting your creative process in a way that proves you're the one pulling the levers. Save your notes. Save your drafts. Treat your code as a creative work, not just a set of instructions for a compiler.

If you want to build a better record, you have to be intentional. Stop relying on the assumption that a repo is a vault. A repo is just a notebook that anyone can write in. Your authorship is in your creative choices, and those choices deserve a record that is as human as you are.

Check your current project documentation against the latest Copyright Office guidance to see if you have the evidence you need. If you're concerned about your coverage, read our full methodology on documenting creative labor.