It says in the link you gave :
The researchers explain this result by noting that they were using group velocity to measure the light's speed—a measurement of the group's envelope speed. The mask, they explain, caused some of the photons in the group to move at a slight angle to the other's causing a slowdown for the group as a whole. Thus, their results are not going to upend one of the basic tenets of modern physics, it is more likely that future researchers will have to make sure lab or astronomical observations are not being impacted by shape changes that occur naturally.
Bold mine.
Light is an emergent collective phenomenon from a large number of photons. Photons themselves always have velocity c but different path lengths add up to a slow down of the light they compose.
The interesting part of the observation is :
because it demonstrated that the speed of the photons was slower than normal after passing through the filter—light is supposed to speed back up to its normal constant after passing through a medium. The experiment showed that light can be caused to travel slower than c, by changing its shape.
Bold mine.
And from the publication's abstract :
Our work highlights that,even in free space, the invariance of the speed of light only applies to plane waves. Introducing spatial structure to an optical beam ,even for a single photon, reduces the group velocity of the light by a readily measurable amount.
Bold mine.
The fixed velocity of light in vacuum is part of the classical Maxwell's equations, and in this sense, to find the group velocity of light slower in vacuum because of the shape of the group is a bit surprising. My hypothesis is that this effect is due to the quantum mechanical level of lasers and photons, and cannot be explained with classical waves. In effect one more demonstration of the underlying quantum mechanical nature of light.