I'd probably approach 90% confidence with ~2-3 different measurements from different labs showing similar stories. Select from the grab bag of possible experiments that could all collaborate and provide consistency in their stories: specifically things like measuring the critical current, critical magnetic field, high quality electrical resistance in various ways (lots of transport), STM measurements of gaps, ARPES measurements of gaps, XRD showing crystal structures are what we expect that matches the DFT, a study on the domain sizes/effects (since this is relying on a substition at the right place, necessarily I suspect that there will be domain effects / "doping" style things... so how those play out in terms of purity and quality of results will be interesting). There will be a lot to study and characterize, and every extra measurement that is consistent rapidly increases that confidence. I'd say I'd get to "100%" (as close as someone could be) after ~10-20 papers across many different labs and measurements. I'd hope we'd get there in ~3-6 months.
What I'm actually even slightly more excited about is "what comes next" -- not the market part, but the "fast follow science". For instance, in the few years after LCBO and LSCO were found (TC~30K) we quickly found YBCO and BSCCO (Tc~130K). I would expect that we'll find a whole class of these materials with substitution tricks that possibly work, and there will be a whole slew of options for "going to market" with the technology. The door this opens is what is more exciting than the specifics of LK-99 itself in my opinion.
Estimating times, after the fast follow science (0.5-2 years optimistically?) we will hopefully have the actual "we're all convinced this is real, and the technology can start to be applied in real devices". Specifically, after everyone is pretty clear on a lot of the material properties and ways to reproducibly make high-quality crystals, so consistency is clear on measurements... then begins the cycles on how to manufacture high enough quality material at scale that it can actually be applied. (specifically, these materials (assuming they're like YCBO/BSCCO) are superconducting crystals that have grains, alignment issues, are physically brittle, have homogeneity issues, etc.) While each solvable, these are all real engineering and material challenges that increase cost to manufacture, and all of this will probably take time before we suddenly get wide-scale products that use this (this is all assuming it's real, haha, there's still plenty of reason to be skeptical).
I think a funded and forward-thinking lab with enough postdocs would be trying to grind out a version replacing lead with gold, per the DFT calculations
Few will remember the million replications, but whoever describes a novel version - even if it's no better - will find themselves on equal footing for helping to understand what's happening
The nitter video could literally be a piece of iron, even a ferromagnetic object will turn like that under the presence of a magnet. It certainly isn't proof that the thing they have there is NOT a superconductor, but you can't use that video as evidence of anything.
Iron and ferromagnetic objects would not react that way. The key to this showing diamagnetism is the fact that the same points are repelled in the same way regardless of the orientation of the magnet. If it was ferromagnetic, you could have it stand up on it's tip in one polarity and stand up on another portion of the object in the other polarity, but you couldn't make it stand up on it's tip both times.
What would you need to see to say "I am 100% convinced this is / isn't real"?
And how long would you expect it will be before that occurs?