The ability of tau aggregates to recruit and misfold monomeric tau and propagate across brain regions has been studied extensively and is now recognized as a critical pathological step in Alzheimer's disease (AD) and other tauopathies. Recent evidence suggests that the detection of tau seeds in human samples may be relevant and correlate with clinical data. Here, we review the available methods for the measurement of such tau seeds, their limitations and their potential implementation for the development of the next-generation biomarkers.