Most startups that fail wrote code extremely quickly and have a cobbled together system that the devs would be ashamed to open source. There may even be private

Most startups that fail wrote code extremely quickly and have a cobbled together system that the devs would be ashamed to open source. There may even be private keys and sensitive info in the code base itself. Additionally if the failure was an acquihire, or a sale of the business assets to a competitor then the code base probably belongs to someone else now, even if it will never be used or see the light of day again. Even if the code is in a state where it could be open sourced legally and safely it would probably be next to useless to anyone but the devs that wrote it. This is because most startups don't tend to write proper documentation, because their small teams can rely on institutional knowledge. (Unless the startup is extremely professional with some truly top notch engineers powering it the documentation effort tends to start sometime after the startup scales to a mid sized team and now requires it.) But when a startup dissolves all that brain space info about how the service works, the endpoints, etc disappears. What you are left with is an undocumented mess that could probably be rewritten from scratch better and faster.