The widespread misuse of drugs has increased the number of multiresistant bacteria1, and this means that tools that can rapidly detect and characterize bacterial response to antibiotics are much needed in the management of infections. Various techniques, such as the resazurin-reduction assays2, the myco- bacterial growth indicator tube3 or polymerase chain reaction- based methods4, have been used to investigate bacterial metabolism and its response to drugs. However, many are rela- tively expensive or unable to distinguish between living and dead bacteria. Here we show that the fluctuations of highly sen- sitive atomic force microscope cantilevers can be used to detect low concentrations of bacteria, characterize their metabolism and quantitatively screen (within minutes) their response to antibiotics. We applied this methodology to Escherichia coli and Staphylococcus aureus, showing that live bacteria produced larger cantilever fluctuations than bacteria exposed to anti- biotics. Our preliminary experiments suggest that the fluctu- ation is associated with bacterial metabolism.