The ELISA test is used to screen blood samples for antibodies to the HIV virus. It measures an "absorbency ratio." Blood samples from HIV-infected donors usually, but not always, give high ratios, while those from uninfected donors usually, but not always, give low ratios. Consider using this test on a sample of blood. The null hypothesis is that the blood sample is not infected with HIV. The alternative hypothesis is that the blood sample is infected with HIV. In this context, what would a Type I error be?