You don't necessarily have to make false promises. You can be totally honest and upfront that the functionality doesn't exist yet and needs to be built first,

You don't necessarily have to make false promises. You can be totally honest and upfront that the functionality doesn't exist yet and needs to be built first, but that you think you understand the problem space and can handle the engineering, provided you can secure the necessary funding, where, by the way, getting a contract and some nominal revenue now could greatly help make this a reality... And if the upside sounds convincing enough, a potential customer might happily sign up to cover part of your costs so they can be beta testers and observe and influence ongoing development. Of course it happens all the time that the problem space turns out to be more difficult than expected, in which case they might terminate the partnership early and then the whole thing collapses from lack of funding.