Orion has been launched once, on a short 4-hour test flight, during which it only orbited the Earth twice. That spacecraft is very, very different than the Orion that is scheduled to launch with humans inside in 2023. It's not a matter of being "willing to accept the risk" - it's just impossible. Safety is one of NASA's highest priorities, (if not the highest, especially after Columbia), so even considering doing such a thing would be unthinkable and is a non-starter. Even if you manage to somehow sidestep decades of a deeply ingrained culture of safety, it's just not even a possibility right now - significant portions of the spacecraft's design have yet to be finalized, and then you have to figure out how to manufacture it, and once you've actually manufactured it, it goes through several rounds of insane amounts of testing and revision before getting anywhere near the launchpad. Spaceflight is incredibly complex - even if NASA somehow managed to get a blank check (like it did during the early space race) there is still a very significant amount of work to be done that takes a very long time, no matter how much money you throw at it.

Aside of the fact that that was only a very early prototype, the Orion service module is actually not a NASA project – it’s actually contracted out to ESA.

And a crew module without a service module is not good for much.