Acquired hemophilia A (AHA) is due to autoantibodies against coagulation factor VIII (FVIII) and most often presents with unexpected bleeding. In contrast to congenital hemophilia, the patient's residual FVIII activity does not seem to correlate with the risk of bleeding as suggested from previous studies. Risk factors for bleeding have not been described. We used data from the prospective GTH-AH 01/2010 study to assess the risk of bleeding and the efficacy of hemostatic therapy. FVIII activity was measured at baseline and weekly thereafter. Bleeding events were assessed by treating physicians. A total of 289 bleeds was recorded in 102 patients. 141 new bleeds starting after day 1 were observed in 59% of the patients, with a mean rate of 0.13 bleeds per patient-week in weeks 1 to 12, or 0.27 bleeds per patient-week before achieving partial remission. Weekly measured FVIII activity was significantly associated with the bleeding rate, but only achieving FVIII ≥50% abolished the risk of bleeding. A good WHO performance status assessed at baseline (score 0 vs. higher) was associated with a lower bleeding rate. Hemostatic treatment was reported to be effective in 96% of bleeds. In conclusion, the risk of new bleeds after a first diagnosis of AHA remains high until partial remission is achieved, and weekly measured FVIII activity may help to assess the individual risk of bleeding. These results will help to define future strategies for prophylaxis of bleeding in AHA.