GUT does not unify gravitation; that would be a Theory of Everything (TOE). GUTs only seek to unify the strong and electroweak forces. TOEs are even more speculative than GUTs. AFAIK there are no purported GUTs which do not use gauge groups, so I don't understand your point about "force".
> In QM ... gravity doesn't exist
The Standard Model of Particle Physics, a particular relativistic quantum field theory, is Lorentz-covariant by design (and with good reason; Lorentz symmetry is a highly-tested feature of our universe). General Relativity as a physical theory of our universe is defined on a Lorentzian spacetime. Where the radius of curvature is larger than the Compton wavelengths involved, adapting the flat-spacetime Standard Model to curved spacetime is straightforward textbook stuff. (Wald 1995 ISBN 0-226-87025-1; Birrell & Davies 1982 ISBN 0-521-23385-2 and several others).
For smaller radiuses of curvature, of if the semiclassical approach is unsuitable because of nontrivial superpositions, one can take a second-quantization approach ("CQG", very definitely a QM theory in which gravity "exists" as a quantum field with potentials and dynamics, and which admits the correspondence principle) or https://webspace.science.uu.nl/~hooft101/lectures/erice02.pd...> both of which match General Relativity as effective theories (the characteristic energy cutoffs are usefully large and/or the characteristic cut-off lengths are transplanckian) are verified to high precision in manifestly relativistic systems (e.g. triple PSR J0337+1715). Are these "wrong"? Prove it. You'll likely need to look into a region of strong gravity, and for that you have to deal with a cosmic censorship conjecture.
> General Relativity is wrong.
Prove it?
I'm with Clifford Will https://en.wikipedia.org/wiki/Clifford_Martin_Will> on this; penultimate paragraph of his 2020 book with Nicolas Yunes ISBN 9870198842125, https://global.oup.com/academic/product/is-einstein-still-ri...>: "It would not surprise me if the solution to the universal acceleration turned out to be simply Einstein's original cosmological constant, a constant of nature like Planck's constant or Newton's constant of gravitation ... And it would not surprise me if general relativity turned out to be absolutely correct according to any future experiment accessible to humankind." The authors' following paragraph is a challenge to you which is not met by your bold assertion at the top of your comment.
General Relativity may be wrong, but that is not proven. Will would be happy to be surprised by it being proven wrong experimentally. (A huge amount of his research has been aimed at what exactly it means for General Relativity to be wrong).
What I would agree with is that we do not know at present how to avoid divergences of the curvature scalars in the limit of strong gravity, and that this means we can't always fully solve a whole spacetime from initial values. But we also don't have any idea how to test whether that actually happens in our universe (that such divergences occur in stellar remnants or at the centres of galaxies is untested hypothesis) so GR is a perfectly sound, mathematically complete effective field theory at the very least.
And moreover the answer may come from new physics in the matter sector, rather than new gravitational physics. What does the Standard Model say about particle energies above YeV? Discovering more about the mechanisms that generate the stress-energy tensor is not the same task as changing the theory of General Relativity.
Meanwhile, there are hundreds of pages in Will & Yunes's excellent book which explain to lay readers the match between GR and observation and experiment, and a wide body of specialist literature on a century's worth of experimental and observational tests of General Relativity, every single one of which has supported the theory or was inconclusive.