I have to wonder if the same issues apply to xylitol. Both erythritol and xylitol are polyols (sugar alcohols) but they're absorbed quite differently. Erythritol is absorbed like sugar but builds up in the blood as it isn't metabolized and is excreted slowly. Xylitol isn't absorbed at all — in humans; it’s quite toxic for many other organisms, including our pets — and passes out the stool. I’ve seen a number of studies showing that it’s healthy for the microbiome — but that means it can give you gas, which I take to be the reason erythritol has been preferred for factory ‘food’.

As a side-note, there’s erythritol in a number of fruits, fermented foods, and blackstrap molasses. Those are all GRAS (generally regarded as safe) so there’s clearly a dosage relationship (i.e. ‘the dose makes the poison’).

Then there’s allulose (D-psicose), the non-sugar sweetener I personally think tastes closest to table sugar. It’s an epimer of fructose (chemically the same but shaped differently because of how the bonds are arranged), so sweet, poorly metabolized, but doesn’t seem to accumulate in the blood.

Like erythritol, it’s present in some fruits (raisins and dates) and considered GRAS by the FDA — but IMO it hasn’t been in general usage long enough for epidemiology to note any negatives. I’ve used it for cranberry sauce as we have one diabetic guest for Thanksgiving. If you’re about to do the same, the flavor is identical (sugar replaced 1 for 1), but it doesn’t gel the same as with sugar. I’ll add pectin next year.

Also, I’ve seen at least one study (in rats) in which adding psicose to the diet favorably alters blood lipids — and another in which it reduces obesity. Rats, though, so it could be different for humans…