I'm rather new to bioinformatics, so this might be a rather basic question, but what are sequencing artifacts, in the context of variant calling?
Artifacts are variations introduced by non-biological processes. Sequencing involves a lot of chemical reactions, and changes in nucleotides that can seem like mutations are really introduced somewhere in the processing of the sample. That's what we call "sequencing artifacts". Proper experiment design, with careful replication, should help you clarify which changes in nucleotides are biologically significant.
Some specific artefacts are:
apparent insertions or deletions in the sequence, which are really just skipped base or duplicated base errors during the sequencing
apparent variants which do not exist in the sample but are introduced in the data because of miscalling at individual bases
differences in abundance of variants between the output and the sample due to biases in PCR amplification, or biases in the ability of the sequencer to sequence each variant, or due to stochasticity in abundance of molecules in the sample