Open Source, Licensing & AI-Assisted Development

Two forces shape how software is made today. Open source — freely shared code that runs much of the world — and AI-assisted development — coding alongside AI tools — are both changing the day-to-day work. A beginner should know what each is, and the rules and limits that come with them. (AI tooling is a fast-moving field worth following on your own; here we cover what it is, not how to wield a specific tool.)

Both are about building on others' work — open source on shared code, AI on patterns learned from vast amounts of code — and both come with responsibilities.

A pair of pictures: open source is a vast shared public library, where borrowing is encouraged but each book has its own rules; an AI assistant is a fast, confident intern whose work is genuinely helpful but must always be checked.

Open Source

Open source is code that's made publicly available for anyone to use, study, change, and share. An astonishing amount of the software world runs on it — operating systems, languages, the libraries from Chapter 6. It's collaborative on a global scale: people and companies contribute fixes and features to projects they all depend on. Open source is less a niche and more the foundation nearly all modern software is built on.

Licenses Matter

"Free" still comes with rules, set by the project's license. Permissive licenses (like MIT or Apache) let you do almost anything, including using the code in closed commercial products. Copyleft licenses (like the GPL) require that if you distribute software built on them, you share your changes under the same terms. Picking up open-source code without checking its license is a real legal risk — the obligations are genuine, even when the code is free.

AI-Assisted Development

AI-assisted development means coding with help from AI tools that suggest, explain, and generate code as you work. They're genuinely useful — drafting boilerplate, explaining unfamiliar code, suggesting fixes — and they're now a normal part of many developers' day. But they're assistants, not authors: they're confidently wrong sometimes, don't truly understand your system, and shift the work toward judgment and review rather than replacing the developer.

Using AI Responsibly

The golden rule with AI tools is: you're still accountable for what ships. That means review everything the AI produces as carefully as a human's code, watch for licensing and security issues in generated code, and never paste in code you don't understand. AI doesn't remove the need for the discipline in this whole course — testing, review, design — it makes that discipline more important, because now there's more code to vet, produced faster.

Cadence is built on open-source libraries — the team checks each license is compatible with their product before adopting it — and the developers use an AI assistant to draft boilerplate and first-pass tests. But every AI suggestion goes through the same pull request and review as human-written code: Lena reads it, questions it, and only merges what she understands and trusts. The AI speeds them up; the discipline keeps them safe.