Hacker News Reader

Comment by jandrewrogers | Hacker News Reader

It is damn near impossible to search on Google for this literature today. Fortunately, some of the links have been posted to this site before, which is searchable. :)

Here is the first good reference I could find, which surveys some of the other literature. It mentions lead in rat models.

https://pmc.ncbi.nlm.nih.gov/articles/PMC2246629/

View on HN

Since maybe half a year ago, the best way to search stuff like that, by a huge margin, are frontier models like ChatGPT or Gemini. Here's what they found, the TLDR is that lead is much less likely to be essential (in extremely minute amounts) than arsenic, which has been proven to be beneficial in trace amounts for a bunch of mammals. Since those experiments cannot be done on humans, we don't know if it is the same situation in our case, but the null hypothesis at this point is that it is.

https://chatgpt.com/share/6a0d522f-5688-83eb-b80c-0f7b157fae... https://gemini.google.com/share/e2e52489a41e

That is an accurate summary of the literature as I remember it. Arsenic has been replicated in a bunch of animals including non-mammals. It seems to generalize well and the background is substantive. There are also plausible hypotheses for mechanisms of action. The assumption is that a common effect across a sufficiently diverse set of animals applies to humans, which is probably a good heuristic.

The evidence for lead is much more sparse. It is interesting, and plausible, but without more evidence it is more of a curiosity. For better or worse, there has been zero interest in investigating that anomaly further. People should avoid lead exposure regardless.

Thankfully elements like mercury never show up in these studies. Nothing in my chemistry training suggests that mercury by nature could ever be anything but toxic to life and the evidence seems to support that. It would be weird to find out that was wrong, though I would accept it with sufficient evidence.

Reply on news.ycombinator.com