Aha, thanks for the link. It's sort of interesting but it does not conclude that Planck's Law is wrong.
What it is doing is testing the
limits of applicability of the law, which is something rather different. As I said, the law was formulated for an ensemble. But it is also formulated on the basis of an assumed relationship between frequency and the number of oscillators, which works very well for macroscopic objects. What they've done here is make an emitter and an absorber that are actually smaller than the wavelength of the radiation frequency range they are interested in. And they have found a resonance phenomenon that
couples the emitter and the absorber to the radiation, such that it increases the rate of emission and absorption relative to what Planck's Law predicts. (They don't give enough details in the press release for me to understand what the physics is behind this, but I'm guessing phonons will be involved somewhere along the line, as it is IR radiation and the vibration modes of crystalline solids are often treated as quantised by means of phonons. My guess is what they have done is a bit like producing an emission and absorption
line spectrum, rather than the usual black body frequency
continuum curve. But I speculate and I'm not a physicist.)
None of this means any textbooks have to be thrown away, or that any application of Planck's Law in other fields of science has to be reappraised. All it means is that there is an interesting phenomenon here that someone may one day dream up a use for.
But thanks again: I learn something new every day!
(P.S. If anyone with a physics background can comment further on this I'd be intrigued to understand more about it.)