A well built amp yes. Having the proper biasing and filtering and VSWR protection, and maybe some other stuff. Low drive, no addition of any driver, not even within the amp, like a 1x4.
Carrier output would obviously be whatever RF power is placed into the amp.
All this as well as good input/output tuning, and like I said, anything I may have left out, would be the way to go. There are other trsnsistors that can be used and aren't dangerous as everyone thinks. At 50vdc remember all you are doing is multiplying the voltage to create a need for less amperage. So 12vdc at 300a, would only be a lot less amperage at 50vdc. I truly have worked on both 24vdc and -48vdc as well as 48vdc systems. Installing power plants and batteries. Then running all power for equipment and such. Both are dangerous, but aren't anymore dangerous than a high amperage 12vdc system. I will leave that at that. Yes it would take some DC to DC conversion, but if you look around for commercial stuff like CK mentioned, it can be easy to find, and probably for cheap at the rate that many communication systems upgrade their Power Plants and Batteries truly at a remarkable rate. Anyways, me not being an amp tech, but knowing some about voltage and amperage, I would think that building a nice unit like this is totally possible and there are some builders that are already working with other things like LDMOS stuff. At any rate, don't get me wrong DM makes a nice amp, but they are more for keydowns and AM use mostly.
Unless you have the bias added and have filtering and such added, like said, they will splatter like hell with the drive that was being delivered to the amp in the video anyways. Now with a proper bias circuit with thermal tracking like stated, along with an Addendum stating to at exactly what drive level is needed to get the carrier approximately where it needs to be. At least something in wrong saying, start with .1w and slowly move carrier up to reach adequate carrier. Also pep power should be true and not some marked up "inflated" numbers at some ridiculous voltage. Most people run 12vdc systems that charge at 13.8-14.2vdc. This should be the voltage that the wattage numbers should reflect. There is much more that I could go on about that could be simply printed into an instruction manual. Haven't seen one of those yet with any 11m amps. Or maybe I just didn't get anything as such due to the fact I bought my gear used. At any rate, there are better ways to obtain clean wattage with low input drive. A simple single final radio like a uniden 980 with adjustable RF power added to it would be a great radio for this. Doesn't have much power, but is clean radio.
At any rate, yes staying within the manufacturers stated power levels would be the way to go for sure. Don't over drive the DM amp into a total splatter box and it will work on am. But without the driver and all that jazz. At least the one in the video, as I beleive this is the one we are speaking of. The new 400 correct. 4 transistor amp.
I run my TNT 600 with my uniden 980 and it truly does a great job, even on SSB I get no complaints from anyone. I have never really put more than 40-50w pep into the amp on SSB, and it's an older one with toshiba 2879's (red dot), transistors.
With that being said, I can get 750 watts out of it with about 50-60 watts of drive. And it's got heavy attenuation (padding), for a high pep input.
But heck with the 980, it does just as well, although I might loose a couple hundred watts, it's not enough to worry about. The uniden with the DK set at 4w and swinging to about 10-12 watts pep on AM using a radio shack meter, it keys up at about 80-100 watts if memory serves me right, and will swing to about 450 watts or so on the radio shack meter on SSB. Take into account that this isn't a true peak hold meter and I would say the amp is closer to 500 watts pep on SSB. Again, nothing but clean reports from locals to DX. And some critics I promise. But never was told I sounded bad one bit. Anyways. The DM amps are nice, but if I was in the market for a new amp, I would be looking into someone that is messing with LDMOS stuff. JMHO. Have a great day all.