To get as basic as possible, an iPhone 5S has a ~6 Wh battery that Apple claims will operate the phone for up to 250 hours on standby. So, if your phone is sitting on your desk doing nothing in an area with good signal, it will require a mere 6/250 = 0.024 Watts just to keep the battery from running down further than it already has.

Delivering even this tiny amount of energy to a phone-sized traducer that may be tens of meters from the emitter and oriented randomly is going to be very difficult. An omnidirectional emitter would likely require more power than a Megadeth concert and heaven help anything with a millimeter scale resonant frequency that's in the room! Tracking the phone's position and delivering a tightly focused beam is probably the only realistic way to go about this. That means you will need some very cutting edge focused ultrasound beam transducers (not cheap) that can mechanically track phones (not cheap) which must be pointed by something like a kinnect (not cheap) and a clear line of sight to the phone (completely unlike WiFi). It's probably going to have to let the user know when there isn't a clear LOS too, because it would suck if your phone died because you set it down behind a plant.

I can't say all this is outright impossible to do at a competitive price. Danny didn't convince me that it's impossible. It is probably pretty close though.

So, why is this company being funded to do something that's probably impossible? Well, Danny is right about one thing: Investors often invest in impossible things. Just google "over unity" generators (better than perpetual motion devices basically) if you don't believe me. Earlier this decade Steorn suckered millions of euros out of investors with a lot of talk and a few cheap parlor tricks. UBeam might not break the laws of thermodynamics, but Steorn literally scoffed at them. People still lined up to invest. Whether Steorn and his co-workers were/are truly insane or con-artists has not been determined to this day. Their website is still live so, against all reasonable expectations, Steorn is still viable!

Personally, if I wanted to scam investors I'd choose something that has a direct impact on the layman's life and theoretically has the potential to be big. Simultaneously, it would not obviously break the laws of physics but would be difficult enough for a company to spend years working on it only to fail. UBeam would be a pretty great setup for a scam actually.