Every now and then, there's a news article where they talk about something that might be interesting for astronomy. There was the meta materials article where they talked about achieving, in the lab, a substance with a negative index of refraction. The work was done in microwaves, which are a good deal longer than visible light. So, although a visible light version would be much, much harder to do, there wasn't anything in physics that suggested it was impossible. And the interesting thing is that such optics might be able to produce images with a resolution limit ten times smaller than the diffraction limit we normally get. Even if it were incredibly expensive, it would make a 3 meter space telescope act like a 30 meter space telescope. But though it's been years since this announcement, i've not heard anything more about it.
So here's another idea. Read this article. I'm not going to repeat the article here. The idea is to sort of combine lucky imaging with adaptive optics. Take lots of images, but treat the pixels independently, to improve the wavefront of the distorted light. You're still using today's massive compute power to make the final image work. And, the article says to expect real results in months.
I've done estimating before. Take the original reasonable estimate, double it, then change the units up one. So if they say 5 years, then it's doubled to 10, and instead of years, you have decades. But if the estimate is 2 months, double it to 4 months, then change months to years - 4 years. This estimating technique is scary on how accurate it is. And though 4 years sounds like a long time, it's not. This is very exciting.