HLI (10k WGS paper) Extented Confidence Regions Positions for SNV analysis

Question

[QVINTVS_FABIVS_MAXIMVS]

Recently Venter's newest toy published 10k whole genomes in PNAS. In the publication the authors mention extending Genome in a Bottle's high confidence regions to an "Extended Confidence Region" comprising 84% of the genome.

I'm interested in rare variants, thus I posit I should limit variant calls to these regions (right?). I assume for GiaB I can take the union of the high confidence bed files in all the samples (right?)

But I would like to use HLI's data. Does anyone know where I can find the positions of the Extended Confidence Regions?

The author's wording in the paper makes it seem that it is available, however I'm less certain.