Stars are colorful but the rods in the human eye that help process very low light intensities aren't color sensitive. The colors of the stars, though, are real, in that if you had enough brightness you'd see the colors with the cones in your eye.

If you put a very colorful picture on the ground but illuminated it with only faint moonlight, your eyes would struggle to see the colors on the picture, despite the fact that the colors on the picture are objectively real, with real pigments that actually represent colors that would be seen you had more light.

So merely pointing out the deficiencies in the human eye don't actually prove that the colors aren't real.

It's that way with Webb since it focuses on infrared, but I thought hubble used the visible spectrum.

After a brief search, it looks like Hubble uses the entire optical spectrum which includes some IR and UV along with visible. It depends on the specific image, but the deep field stuff looks like it was a combination of visible and IR, which makes sense considering red shift. But the bluer objects were captured in the visible.

So they inevitably had to compress the spectrum for the photos, but speaking as somebody who has taken tens of thousands of photos in RAW format, all the colors in every photo are translated data. :) (that also goes for the screen displaying the final image using a mix of three wavelengths rather than the actual colors of the original light)