I like the following analogy to explain how this experiment could proceed. It's very simple if thought of in these terms and very complex otherwise:
An architect builds an office complex with several buildings in the center of a field. No parking or walkways to each building are designed or provided. Over the course of several months a beaten path develops and patterns are made visible. The path becomes a dirt trail and eventually is paved.
Risk cannot be managed effectively by market forces. It requires knowledge and wisdom, acquired through experiences, failure and success. The conservative investor will not participate in a DAC where he or she funds this discovery process. Only the benefactor, benefited or gambler would invest.
The burden on the adjuster and investor is too great without substantial predictability. I believe a third actor is required, an "oracle". Ideally the oracle: is developed and then self evolves without intervention. The oracle(s) could initially be semi static repositories, eventually self correcting.
The oracle might then hone its power to assess risk. Pools of varying risk would organically form over time, investors could then decide when, if and how to jump in. Some pools will be too shallow and fail to attract swimmers while others could prove to be popular and become overcrowded.
While being quite possibly the most lucrative proposed DAC, it's certainly the most complex I've seen envisioned with the most unknown factors, ergo the need for an oracle(s). For a taste of the complexity, google "ruin theory". it's way over my head and thoroughly humbling.
Btw, As I wrote this post, I felt as though I was channeling Peter Sellers/ Chancy from the movie "Being there". First comes spring, then summer, fall, winter then spring again .