Sure. A genome-wide perturb-seq experiment is a huge (and expensive) technical accomplishment, but the authors did not "map every human gene to its function". (Nor did they claim to, it's press release hyperbole.)

One, there's ~20k protein-coding genes in the human genome, and they screened ~10k, analyzing about 2k (fig 2a).

Two, all the functional annotation is based off transcription profiles. They essentially looked for clusters of genes with correlated expression, and assigned function based on genes with previous annotations (fig 2d, S4).

It's a good resource, but there's a lot more molecular work to be done to validate the function of these genes.