Cool idea, very ambitious, is there any prior research or feasible testing setup that would support getting from 0 to 1 with this?
Yeah this seems crazy to just throw out into the public. How do they decide what and when is worth funding? Does there have to be some kind of shining light of progress... or is it just moonshot after moonshot of blind hope?
PMs are typically experts in their field and often come from an academic, industry, military, or government research lab. DARPA gives them a budget for grants, autonomy, and 3-5 years to fund interesting work. Many fail, many have mixed success, and a small number revolutionize their field.
(Source: worked on a few DARPA-funded projects decades ago.)