Microbiologically influenced corrosion inhibition (MICI) has been observed for Al 2024, mild steel and cartridge brass when an artificial seawater (AS) solution containing growth medium was contaminated by bacteria. In this case the test electrodes had been sterilized by immersion in ethanol, while the test cell had been sterilized in an autoclave. In tests in which all parts had been sterilized in an autoclave contamination did not occur and the corrosion behavior was similar to that obtained in AS without growth medium. Impedance data clearly showed that pitting did not occur for Al 2024 in the contaminated solution, while for mild steel and brass corrosion rates were greatly decreased. The corrosion potential E-corr decreased due to bacterial contamination for Al 2024 and brass, but increased for mild steel. Analysis of electrochemical noise (EN) data demonstrated that the power spectral density (PSD) plots of the current fluctuations for Al 2024 and mild steel changed to lower levels due to bacterial contamination, while the PSD plots for the potential fluctuations remained unchanged. The slopes of the potential and current PSD plots were unchanged, while the localization index (LI) was close to zero in all cases. Pitting of Al 2024 did not occur and corrosion rates of mild steel and brass were greatly reduced during exposure to AS containing one of two strains of Shewanella. The corrosion mechanisms seemed to be different for mild steel, where E-corr increased in the presence of the bacteria, and Al 2024 and brass, where it decreased. Addition of kanamycin to AS containing S. ana after 7 days induced an immediate increase of E-corr for Al 2024 and brass. However, the corrosion activity occurring in abiotic AS was observed only after about another 7 days.