Under no circumstances do you "gotta hand it to them", but Musk took the private beta, US-only Birdwatch, renamed it to something less cute (but ultimately a more pragmatic name), and launched it more broadly.
I think the point is that while he doesn't really deserve credit for developing the project, he was in charge when it was rolled out. He could have easily killed the beta project as having failed or a leftover project from the people that used to be there.
Sure, he can get credit for not making a bad decision in this case and defaulting to a good outcome. (He met the minimum benchmark a product manager would be expected to do, but let's not overstate his part)
When it comes to a product decision, those are very different in my book. The default position should be not to change or intervene. In the case of a new product or feature that means not shipping the product unless it is tested along the way and continues to meet predefined targets. Anything less risks committing resources to a project that at best isn't valuable and at worst is a net drain on the company.
In this case, Twitter after Elon acquired it decided to continue with the beta feature. I have no idea how that decision was made internally and can't peg it on Elon, though I can assume he had the authority to spike the project if he wanted to. Assuming Twitter has competent PMs, they made the active choise to move the project out of beta when the testing phase was complete and they evaluated whether it met goals.
Alas, twitter is has never been like "every single company", and I presume it would have languished in it's pre-musk state for many more years before it got cancelled.
I don't see any reason to think so. As an early Birdwatch/CN contributor and also a long-time user of the API, I've always thought their back end stuff was pretty great. API v2 also rolled out pre-Musk and was actually great until he crippled it with absurd limits. I'm not going to pay thousand$ for the privilege of churning out free network analysis; last time I checked you could only collect 1500 tweets/month before a very expensive meter started running on your API calls.