Infections due to Cryptococcus neoformans cause severe disease, mostly in AIDS patients. The antifungal drug recommended for the initial treatment of these infections is amphotericin B with or without flucytosine, but treatment failure occurs, associated with high mortality. Thus, antifungal susceptibility testing is needed. However, the in vitro susceptibility tests available for C. neoformans are not useful to detect isolates that are not susceptible to antifungal agents such as amphotericin B. The aims of the present study were: (1) to determine and compare the in vitro activity of amphotericin B against C. neoformans clinical isolates by using different dilution and diffusion methods; (2) to evaluate the concordance among the methods used and the reference method; (3) to evaluate which method could be the best to correlate with the clinical outcome. The reference method EDef 7.2 from the European Committee on Antimicrobial Susceptibility Testing and commercial Etest strips were used to determine the minimal inhibitory concentration against amphotericin B. curves, minimal fungicidal concentration, and a disk diffusion method were also developed to evaluate the cidal activity of amphotericin B. The time-kill curve assay showed correlation (p < 0.05) with clinical outcome, whereas EDef 7.2, minimal fungicidal concentration, Etest, and disk diffusion showed no correlation (p > 0.05). Thus, the time-kill curve assay could be a potential tool to guide a more efficient treatment when amphotericin B is used.