Using the known numbers and evolution of unobscured AGN, and assuming that there is a large number of corresponding obscured AGN -- as expected if AGN unification is correct -- we predict the infrared number counts of AGN. These agree well with the observed GOODS Spitzer data at 3.6, 8 and 24 $\mu$m, provided that the dust content of the obscuring torus is $\sim 2$x less than assumed. This strongly supports the AGN unification picture, and implies a large population of infrared bright obscured AGN at all redshifts and luminosities.
Categories: astro-ph