That was interesting. My go to review on the subject does include a description of GR Casimir force derivation and adds the Lamb shift, and like the article reference from Jaffe concludes that quantum zero energy of the vacuum are real. [ http://arxiv.org/pdf/1205.3365v1.pdf ]

Now I note that the review claims the difference between the vacuum energy density estimate and the measured value is something like 54–56 orders of magnitude off, not the naive 121–123 oom. I wondered why, and it seems they use renormalization to make the result compliant with relativity. [Lorenz invariant; https://arxiv.org/pdf/1105.6296v1.pdf ].

That looks at first like finetuning to me [not knowing either QFT or GR]. But they claim in the appendix that the renormalization removes a divergence in a way that doesn’t depend on the procedure, precisely when they impose Lorentz invariance. My tentative reading is that the condition therefore looks natural. More handwavingly, maybe gravity/cosmology plays nice (nicer) with quantum field theory than at a first look?

When the article relates Casimir force to gravity (and cosmology, “effective boundaries”), I assume the Lorentz invariance condition in the vacuum energy analysis becomes a must. Interestingly, if supersymmetry also exist it would halve the traditional ~ 120 oom vacuum energy density estimate if I understand correctly, and perhaps that would bring the total estimate down close to the measured energy density!?