The original blog post for Mythos did lay out this safeguard testing strategy as part of their plan.

This seems needlessly cynical. I don't think they said they never planned to release it.

They seemed to make it clear that they expect other labs to reach that level sooner or later, and they're just holding it off until they've helped patch enough vulnerabilities.

My guess is that it is just too expensive to make generally available. Sounds similar to ChatGPT 4.5 which was too expensive to be practical.

It's too powerful now. Once GPT6 is released it will suddenly, magically, become not too powerful to release.