Although viral variability studies have focused traditionally on consensus sequences, the relevance of molecular clone sequences for studying viral evolution at the intra-host level is being increasingly recognized. However, for this approach to be reliable, RT-PCR artifacts do not have to contribute excessively to the observed variability. Molecular clone sequences were obtained from an in vitro transcript to estimate the maximum error rate associated to RT-PCR for the Hepatitis C virus (HCV) E1-E2 region. On average, the frequency of RT-PCR errors was one order of magnitude lower than the level of intra-host genetic variability observed in samples from an HCV outbreak. However, RT-PCR errors were not distributed evenly along the E1-E2 region and were concentrated heavily in the hypervariable region 2 (HVR 2). Although it is concluded that RT-PCR molecular clone sequences are reliable, these results warn against extrapolation of RT-PCR error rates to different genome regions. The data suggest that the RNA sequence context or secondary structure can determine the fidelity of in vitro transcription or reverse transcription. Potentially, these factors might also modify the fidelity of the viral polymerase.