I agree with your general premise that safe spaces are necessary for allowing ideas to develop. I think they’re also necessary for giving people a place to simply exist without their existence being questioned or threatened.
However, I don’t know that I agree with your proposal to avoid echo chambers by encouraging “yes and”-style responses. The sort of person who accepts the core premise of these ideas and wants to improve them can probably already communicate their ideas/criticisms without violating the safe space.
A bad faith actor could always use “yes and” as a means to promote a slippery slope argument that seeks to invalidate the original idea. And there are certainly some ideas that would be dangerous to add on to.
A side note, if you’re interested in discussing how to create a supportive online space, you may want to check out beehaw.org if you haven’t already. They’re a lemmy instance that defederated with lemmy.world, but I believe they’re still federated with your instance. Before they defederated with us, I was able to read a series of lengthy posts by one of their admins describing their moderation philosophy- specifically how it evolved over the years as they tried to create a safe community in other sites and what they settled on for beehaw.org. It was really interesting, and their conclusion was that the only way to create a safe community was to proactively moderate, warning and banning users before they cross a line. Reactive moderation (removing/banning posts and users after they do something offensive) still allows for the negative post to impact the community before it’s removed. They also believe that having clearly defined rules doesn’t work since trolls just learn to work within them while still achieving their goals.
Their moderation style requires a lot of work to implement, but it definitely achieved the results they wanted. The beehaw communities were very friendly and users seemed more open with each other.