In Alzheimer’s disease (AD) and other tauopathies, the cytosolic protein Tau misfolds and forms intracellular aggregates which accumulate within the brain leading to neurodegeneration. Clinical progression is tightly linked to the progressive spread of Tau pathology throughout the brain, and several lines of evidence suggest that Tau aggregates or “seeds” may propagate pathology by spreading from cell to cell in a “prion like” manner. Accordingly, blocking the spread of extracellular seeds with an antibody could be a viable therapeutic approach. However, as the structure of Tau seeds is unknown, it is only possible to rationally design therapeutic Tau antibodies by making a priori assumptions. To avoid this, we developed a robust and quantitative cell based assay and employed an unbiased screening approach to identify the antibody with the highest activity against human Tau seeds. The selected antibody (D), directed to the mid-region of Tau (amino acids 235–250), potently blocked the seeding of human AD Tau and was also fully efficacious against seeds from progressive supranuclear palsy. When we compared this antibody with previously described reference antibodies, we were surprised to find that none of these antibodies showed comparable efficacy against human pathological seeds. Our data highlight the difficulty of predicting antibody accessible epitopes on pathological Tau seeds and question the potential efficacy of some of the Tau antibodies that are currently in clinical development.