The null hypothesis is “no different between A and B”.

Run a statistical test (T-Test) and reject the null hypothesis with a particular confidence etc. I learned this in MTE 201 so my notes should be in the google drive.