Being risk averse is a solid path in situations such as yours. Without a crystal ball it is impossible to reliably predict the potential of an incomplete idea.

Being risk averse is a solid path in situations such as yours. Without a crystal ball it is impossible to reliably predict the potential of an incomplete idea. Perhaps better questions to ask is "What problem does my solution address? How much are potential customers willing to pay?" "Generally desirable" doesn't indicate a real and present demand. With "no immediately obvious market" to me it seems like your "aligned AI agent framework" is a solution in search of a problem.