Blue Monsters. Why are JWST super-early, massive galaxies so blue?

The recent JWST tentative discovery of a population of super-early (redshift $z> 10$), relatively massive (stellar mass $M_* = 10^{8-9} M_{\odot}$) and evolved (metallicity $Z \approx 0.1 Z_{\odot}$) galaxies, which nevertheless show blue ($\beta \simeq -2.6$) spectra, and very small dust attenuation ($A_{\rm V} \leq 0.02$), challenges our interpretation of these systems. To solve the puzzle we propose two solutions in which dust is either (a) ejected by radiation pressure, or (b) segregated with respect to UV-emitting regions. We clarify the conditions for which the two scenarios apply, and show that they can be discriminated by ALMA observations, such as the recent non-detection of the $88\mu m$ dust continuum in GHZ2 ($z\simeq 12$) favouring dust ejection.