The origin of the Binary Black Hole (BBH) mergers detected through Gravitational Waves (GWs) by the LIGO-Virgo-KAGRA (LVK) collaboration remains debated. One fundamental reason is our ignorance of their host environment, as the typical size of an event's localization volume can easily contain thousands of galaxies. A strategy around this is to exploit statistical approaches to assess the spatial correlation between these mergers and astrophysically motivated host galaxy types, such as Active Galactic Nuclei (AGN). We use a Likelihood ratio method to infer the degree of GW-AGN connection out to π§ = 0.2. We simulate BBH mergers whose components' masses are sampled from a realistic distribution of the underlying population of Black Holes (BHs). Localization volumes for these events are calculated assuming two different interferometric network configurations. These correspond to the configuration of the third (O3) and of the upcoming fourth (O4) LVK observing runs. We conclude that the 13 BBH mergers detected during the third observing run at π§ β€ 0.2 are not enough to reject with a 3π significance the hypothesis according to which there is no connection between GW and AGN more luminous than β 10 44.3 erg s β1 , that have number density higher than 10 β4.75 Mpc β3 . However, 13 detections are enough to reject this no-connection hypothesis when rarer categories of AGN are considered, with bolometric luminosities greater than β 10 45.5 erg s β1 . We estimate that O4 results will potentially allow us to test fractional contributions to the total BBH merger population from AGN of any luminosity higher than 80%.