There are expectations that these projects actually deliver their stated goals, including the 'rewards', when they receive the requested funding. It's still a

There are expectations that these projects actually deliver their stated goals, including the 'rewards', when they receive the requested funding. It's still a new experiment overall that can be innovated upon, but it's not immune to a crash in people willing to throw their money at random pet projects. Even if the reality is that most projects deliver, having enough bad apples or outright frauds can be enough to hurt progress. I read a recent piece talking about how many game projects seem to terribly underestimate the amount of money required to fund a game. Enough people who are terrible at estimating their financial needs can also hurt everyone. It's not like this problem can't be solved, but it shouldn't be ignored if you want crowdfunding to succeed.