Microsoft published an AI-generated flowchart to explain GitHub that was a blatant copy of Vincent Driessen's famous Git branching model. The creator called it "careless, blatantly amateuristic, and Microsoft unworthy." They quietly removed it after being called out.
The irony is perfect: Microsoft, which owns GitHub and sells Copilot AI coding tools, used AI to plagiarize developer documentation. This is every creator's AI nightmare - big tech ripping off your work with their own tools, then acting surprised when caught.
Vincent Driessen's Git branching model is legendary in developer circles. Published in 2010, it's been used by millions of developers and became the de facto standard for Git workflows. The flowchart explaining it is iconic - clean, clear, immediately recognizable to anyone who's worked with version control.
Microsoft's AI apparently thought it was a great template. So it generated a remarkably similar flowchart for GitHub documentation, keeping the structure, the visual style, and the core concepts while changing just enough to pretend it was original work. It wasn't.
Driessen noticed immediately and called it out publicly. His response wasn't just about attribution - it was about what this reveals about how big tech companies are deploying AI. "The AI rip-off was not just ugly," he wrote. "It was careless, blatantly amateuristic, and lacking any ambition, to put it gently. Microsoft unworthy."
He's right. This isn't a technical failure of AI systems - they're working as designed, remixing existing content into "new" combinations. This is a failure of process, judgment, and respect for creators. Someone at Microsoft fed an AI a prompt, got back a flowchart, thought "good enough," and published it without checking whether it was derivative work.
Microsoft's response was to quietly delete the flowchart. No acknowledgment, no apology, no explanation. Just remove the evidence and hope everyone forgets. That's not how you handle plagiarism - accidental or otherwise. That's how you handle a PR problem.
Here's what makes this particularly galling. Microsoft sells Copilot, an AI coding assistant trained on billions of lines of public code from GitHub repositories. The entire product is built on the work of developers who uploaded their code expecting attribution and respect for licensing. Now Microsoft is using AI trained on that work to plagiarize documentation, and their response is to delete and move on.
This is the AI deployment pattern we're seeing everywhere: move fast, let AI generate content, deal with problems after publication. It works great for the companies doing it - they get productivity gains and cost savings. It's terrible for creators whose work gets absorbed, remixed, and republished without credit.
The documentation plagiarism isn't even the worst case. At least Driessen saw it and called it out. How much AI-generated derivative work is being published without the original creators ever knowing? How many blog posts, tutorials, documentation pages, and marketing materials are AI-plagiarized versions of someone else's work?
Microsoft knows better. They have processes for documentation review, legal teams who understand copyright, and plenty of developers who could spot derivative work. Those processes failed because no one bothered to check. The AI output looked good enough, so they shipped it.
This isn't about AI being bad or dangerous. It's about companies treating AI output as good enough without doing the work to verify it's not plagiarized, factually wrong, or legally problematic. That works until you get caught, and Microsoft just got caught plagiarizing from one of the most respected voices in developer tooling.
Driessen's work is widely used because it's good - clear thinking, well explained, generous with knowledge. Microsoft's AI ripped it off because that's what these systems do when asked to generate something similar to existing high-quality work. The technology did exactly what it was designed to do. The humans were supposed to catch it before publication.
The quiet deletion tells you everything about Microsoft's priorities. They're not sorry about plagiarism - they're embarrassed about getting caught. If Driessen hadn't noticed or hadn't had the platform to call them out, that flowchart would still be up, and Microsoft would still be benefiting from his work without attribution.
Every creator watching this is getting a lesson in how AI deployment works at big tech companies: your work will be used to train their systems, those systems will generate derivative content, and you'll only get attribution if you're famous enough to make them care. That's not a bug - it's the model.
