The research is kind of hazy. Bisphenol-A has been shown to be a very very weak estrogen when measuring receptor binding affinity (about 37,000 times lower than human estrogen https://pmc.ncbi.nlm.nih.gov/articles/PMC2774166/#sec2), but has also been shown to be a potent stimulator in vitro for specific cell types (https://pubmed.ncbi.nlm.nih.gov/22227557/).
The lowest concentration of BPA that's been shown to be estrogenic according the second article is 0.1pMol/L which is around 230 picograms per litre of blood, or 1.1ng total for an average adult.
BPA's biological half life in humans is up to two to five hours depending on a range of factors (https://pmc.ncbi.nlm.nih.gov/articles/PMC2685842/), so taking the worst case you'd need to be continually exposed to around 2.5ng of BPA over a day.
So 'safe' as defined right now would be keeping the absorption below that 2.5ng per day threshold.
I don't know how how much BPA in plastics can transfer out per day, the research I've seen seems to indicate that unless it's a food container it's pretty minimal but I don't know enough to evaluate the quality of that research.
Your skin is also a pretty good barrier so only around 2.2% of any BPA on your skin can pass through in an ideal situation, so absorption from non-food sources is much lower (https://pmc.ncbi.nlm.nih.gov/articles/PMC9210257/)
The other problem is what do they replace BPA with? To be safer it would need at least as well studied as BPA, but often it seems like the 'safer' options are just not very well studied yet and could actually be worse.