You crowdsource the viability and the platform makes sure they're a real company and verifies their legal status. Who better decide whether or not a company has a viable idea than their potential customers or users? In fact, I can't think of a better way to assess the viability of a company than to put it out there in front of 1000's of eyes to see.

The truth is that the Y-Combinator evaluation style, or other more tradition financing routes for that matter, aren't the end all be all. We're working our tails off to leverage our users knowledge and make a system that roots out any sort of scammers.Crowdsourcing isn't a short cut, it's just a decentralized system versus our current more centralized one. In fact, I suspect more man hours will be put into assessing viability through a crowdsourced model just because of the sheer number of people.

Scammers are going to slaughter you.

Company: "We're going to cure cancer!"

First 20 crowdsourcers, all paid by company: "These guys are totally legit! I visited their offices! Saw a hamster that was totally cancer-free!"

Next 500 crowdsourcers, not paid by company: "Looks legit! Take our money, please!"

Company founders: {fly to Switzerland in their ROFLcopter, having committed no crime whatsoever}

1000 or 100,000 or 100,000,000 eyes looking at a subject shallowly are not a replacement for one set of eyes looking at a subject deeply. The great thing is, the marks won't even realize they've been scammed for years or perhaps ever, due to the long delay between providing money and expecting results.

I think a more subtle approach might work better, like promising a distributed social network written in Rails.