To test the safety of semen two approaches can be applied: checking the end product, or continuous surveillance of the bulls before and after semen production. The first method is examination of semen for the presence of infectious agents. This method depends completely on a single investigation and therefore relies only on the sensitivity of the test method. The second method is testing the bulls for diseases before and after semen collection, based on sequential investigations for the absence of either pathogens or antibodies against infectious agents. The EU-Directive 88/407 prescribes that bulls in AI stations must be monitored for the absence of diseases, but only at 12-monthly intervals, which is a severe disadvantage. Furthermore, the directive is specific neither in the tests to be carried out nor in the specification of some pathogens (e.g. Campylobacter foetus). A programme is presented based on monthly testing of a limited number of bulls for the absence of endemic diseases only, on the basis of Hazard Analysis of Critical Control Points (HACCP). This method only applies to diseases with high transmission rates. Testing some 20% of the animals on a monthly basis can monitor these highly contagious diseases (e.g. IBR). To monitor BVDv infections, however, monthly testing of all negative animals and semen culture or semen PCR of animals that have seroconverted for this virus seems necessary. Endemic diseases with slow transmission rates in bulls do not suit such a system and can only be monitored on an individual basis.