I initially reacted to the root comment with why these kinds of articles often receive dismissive comments. HN is mostly not a medical forum so a typical reader reader isn't going to want nor be able to discuss the technicalities - they just want to know whether to avoid a substance or not. As is often the case, the results are inconclusive, hence the dismissals. (And these dismissals as top comments are also useful for the typical reader as they pretty much want a yes/no verdict and move on.)
But to your points, if there aren't any studies which can show that a compound is dangerous in any meaningful way, why would you want to avoid it? (Given there is a need or purpose, e.g, a low-calories sweetener.)
Also, decisively showing something to be safe is impossible in a similar way that software tests can only show that you haven't found any bugs yet, it doesn't mean there are none. (Off-topic: That's a quote from Edsger Dijkstra, which the following can be added: he is right, but for unit tests - using types, property testing or by running through the entire argument space for a pure function you actually can show that there are no bugs.)
Here's an example. Company A invents compound B and pays company C to do safety studies that monitor the subjects for a few weeks or months. The study shows no significant danger. They start selling compound B in food or as medicine. Then 10 years later after millions of people have ingested varying amounts of compound B, it's found to cause some harm that wasn't found in the initial study. Company A pays a fine of less than the profit they made selling compound B, and compound B is pulled from the market.
There are many stories like that. Should I have avoided compound B on the precautionary principle? Or, because the only science done so far in those first 10 years showed it was safe, should I have considered it safe?
In the case of food additives it's even worse. Company A makes compound B, it's in food, no studies are done, millions of people eat it, are harmed, and no one knows for years or decades.
Personally I think introducing novel compounds to the body is just a bad idea period and I avoid them as much as is practicable. Too many have been found to be dangerous only decades later, and we have a population with rapidly rising rates of chronic disease and cancer etc that could be related to toxic stuff. Why risk it. And especially why trust science that is paid for by the companies that will profit if the science shows their thing to be safe?!
---
There _could_ be ways to show safety of something to a point, like do a 20 (or 50) year study with large cohorts where one population uses compound B and the other doesn't and monitor overall outcomes. But that's far too expensive and time consuming and not required so basically no one does it. Companies want to profit from their novel compounds fast.
Artificial sweeteners replace sugar (as you likely know), so there's a specific reason. But yes, eating any random newly invented molecul for no reason I wouldn't recommend either.
But to your points, if there aren't any studies which can show that a compound is dangerous in any meaningful way, why would you want to avoid it? (Given there is a need or purpose, e.g, a low-calories sweetener.)
Also, decisively showing something to be safe is impossible in a similar way that software tests can only show that you haven't found any bugs yet, it doesn't mean there are none. (Off-topic: That's a quote from Edsger Dijkstra, which the following can be added: he is right, but for unit tests - using types, property testing or by running through the entire argument space for a pure function you actually can show that there are no bugs.)