I don't know if it can be taken for granted that somehow normalizing pressure between the endplates would reduce drag. Yes, drag borne by tip vortices would likely fall. However, increasing static pressure means reducing dynamic pressure (velocity), because total pressure is constant, and reducing the dynamic pressure of air flow between the endplates inherently means reducing the dynamic pressure of air flow upstream, because downstream events have upstream consequences. Such a strategy would create a bottleneck, which is the polar opposite of efficient.
(EDIT: I've parsed that so many times in my head that it no longer appears to be language, much less coherent English. If it doesn't make sense, let me know, and I'll try to clarify.)
Plus, there are no immediate signs of stagnation in the FloVis streaks...
What seems more likely is that the slots effectively increase dynamic pressure - as always.
For the last couple of years, both McLaren and Red Bull have run diffusers that maximize expansion via geometries that make use of every legal millimeter available between the rear wheels...
For comparison, here's SF16-H...
Presumably, the idea is to couple the edges of the diffuser to the upwash created by the brake ducts in much the same way diffusers were formerly coupled to the upwash created by the beam wing. If so, reducing the air flow that passes between the rear wheels and the outside of each endplate by way of full-length slots that reroute such air flow between the endplates would allow for more upwash from the brake ducts, which would then increase efficiency at the edges of the diffuser.
Ultimately, it would allow the rear wing to be set at a proportionally lower AoA with less consequential loss of underbody efficiency compared to the former design. In other words, same downforce, less drag (or any permutation thereof).
At least, that's how I see it.