Worth noting that bulk graphite is probably too conductive (it's fairly metallic as things go); but a thin layer of it, or a mixture, can have higher resistance and thus good absorption.
Graphite is an excellent susceptor at low frequencies, but this is also done with inductive coupling, where the near fields are quite low impedance; impedance in the cavity will be say 100s of ohms, where a higher resistivity material is desirable.
Ideal conductivity is more or less to have a 1/4 wave thick layer and skin depth such that it absorbs in the bulk; or a sheet/foil/film layer that, while lower resistivity, has comparable sheet resistance (i.e. low 100s Ω/sq, I suppose).
Conversely, dielectric absorption approaches from the other way, where impedances might be kohms at lower frequency (various industrial processes that aren't amenable to induction), but converges towards cavity impedance here. I'm not sure what offhand would have reliable dielectric loss though, in the same temperature range, or if it matters much at all how we express the lossiness at this point. (We'd pretty reasonably assume a material like magnetite has a combination of hysteresis and conduction loss, but beyond something like water..?)
Yup -- though graphite burns in air well below 3000C; but at 600C it's pretty slow, and negligible at 450.
From memory, the video applied it with a water glass slurry.
Hmm, interesting catch is, once the glass itself melts (~red hot), ionic conduction takes over. Which might end up making that a bit goopy, I guess? Upshot, it should slow the combustion of graphite, though on the other hand it's also not very sticky, the graphite will tend to float out of it. 450C is probably still okay, but I would be concerned about possible hot spots, I guess?
There's also graphite paper, fiber, foam, etc.; though probably fairly specialty products? (Note, those pyrolytic sheets for thermal conduction and shielding are too conductive, they're in the area of ~ohms I think.)
Tim