This is about how self-limiting contracts and architectures benefit developers by preventing their creations from harming others
We've seen a lot recently in terms of news about tech giants - the ones we use every day - becomming tools of abuse for powerful entities with unsavory agendas, even if their creators would have it otherwise. This is simply because their creations have abilities and information powerful entities want while NOT HAVING anything to keep their technology from being controlled in this way.
We can assume that the founders wouldn't have had it this way, and most founders of these giants aren't really involved in the companies anymore anyway, and who would want to be once they saw they were powerless to prevent their creation being used this way?
There's no one to blame. You can't blame founders or workers for prioritizing their lives and livlihoods over being martyrs for the common good. You can't blame workers within powerful agencies for doing things against their will by harming and abusing technology users for the same reason (except, of course, within the realm of what they can get away with in terms of sabotaging harmful motions without getting fired or worse).
Following this model, we can expect a perpetual repeat of an arc: A brand of a type of technology rises to the point of mass or universal usage. It is commandeered by powerful entities by hook or crook. The abuses of what has now become a malignant tool of powerful entities makes it hated by its users. It collapses. Another brand emerges. Perhaps the owners of the failed brand buy the new brand.
This arc can be avoided at the inception of a new creation by baking in limitations, so that even if a creator WANTED to turn over his technology, it would be impossible.
There are 3 obvious areas for limitations and 2 obvious means of limiting them:
Protections for: user privacy; access to a company's information; ability to be other than transparent with any changes within a company or anything related to user's information.
By means of: baked in transparency on a simple, unchangeable information page/index with records of any changes; user-company contracts that make it impossible for the company to make any changes to original terms or abuse protections without incurring a contract-based lawsuit that would bankrupt and destroy the company.
I don't see how, absent these things, any technology once successful could avoid exploitation by powerful entities and the abuse of its users.