Correct. That means the algorithm should be adjusted to reflect that new learning, or you should adjust your GFs accordingly.
I actually agree with you completely and almost wrote that into my initial response. If the algorithm is corrected, there is no need to have helium content a part of the computer setup.
I do not think that adjusting GFs is a valid approach to using a computer that does not support helium, for trimix diving. Nor, really, for diving deep air. I mean, it is better than nothing. But, I think the "more likely to keep you unbent" approach is to use a trimix computer for trimix diving, and probably even use a trimix computer for deep air, while setting it for some reasonable mix of trimix, even though you're using air. That way, you will get the extra deco needed, even though you're not actually using helium.
Like I said, I do agree that adjusting to lower GF numbers is better than nothing. But, for me anyway, it would fall into the category of "I'm not diving today because I don't have the right equipment."
Regardless, even if the algorithm gets updated (ZHL-16D? Anyone?), I still think there would be a need to have helium content as part of the computer setup. Record-keeping (via downloaded logs), Isobaric Counter Diffusion alerts, and END alerts all come immediately to mind. I think there are several other considerations, as well. The bottom line to me is I think the computer should always be told exactly what you're breathing - even if the deco calculation doesn't care whether there's helium, or it's all O2 and N2 (but I think it would care, even if the algorithm predicted He and N2 to on- and off-gas identically - different partial pressures of each during the dive would mean different tissue loading of each).