There are a few issues with 'safety' that really muddy the waters and you have to decide really how much risk you're willing to take.
1) Concentration is everything. The universal solvent is Dihyrdrogen Monoxide, DHMO for short. In relatively small quantities, inhaled, it can kill you. In larger quantities, ingested, it can kill you. In large enough quantities, it has destroyed entire cities. In certain studies, given the possible effects and damage it can cause, 86% of survey takers voted to ban all exposure to this substance. Every single person who's used it has eventually died. Nearly 300,000 people died from inhaling this substance during 1999-2010. What is it?*
2) Stochastic effects of substances. Some people smoke for 50 years, massively, and do just fine. Others have never touched one and die of lung cancer. All these substances or environmental effects do is change the odds of certain things happening. Statistics do not apply to individuals or individual events. Nevada thrives on people unable to understand this concept.
3) Correlation is not causation. Ice cream sales and shark attacks have a one to one correspondence. It's as perfect as you get in statistics. However, banning ice cream will not prevent shark attacks. There are many, many examples of this in the 'lets scare people into behaving' crew. Diet soda causes weight gain because more overweight people drink diet soda. Insulin causes cancer. (Well, yes. All of those people would have died without it so they lived long enough to get cancer.) And so on.
4) Background and significant change from background. Everyone knows radiation is harmful, right? But, do you know what industry exposes all of its workers to the highest levels?** I'll give you a hint, it has nothing to do with nuclear power, nuclear medicine or even diagnostic medicine (x-rays and the like). The reason you don't see those people dying like flies is because it's still well within the 'noise margins' for normal exposure. Everyone is exposed to background radiation all the time. What makes it directly dangerous is when it gets way outside the normal levels. (As opposed to stochastically dangerous, see above.)
So, you make your choices and take your chances. Myself, I vape a lot of very, very 'risky' things. Cinnamon, heavy menthol, clove, anise, hops, Latakia tobacco extract, if it's 'bad for you', it's probably already in my collection. Very little of what I vape is based on pure, controlled flavoring created by flavor chemists. However, the quantities in the liquid are far, far higher than anything that ends up in the aerosol and what ends up in the aveoli (the tiny air pockets where oxygen is absorbed) is a small fraction of that. In addition, compared to what's in the air, normally, the dust and pollen and moisture and oils and soot and ... it's just one of many, many things out there and not significantly higher than background. And even if it did have an effect and doubled my risk of x cancer. Ok, well, um, yeah, from 0.1 to 0.2, if I don't get hit by a car, run over by a horse, fall off a bridge, drown in the bathtub, ... not worth it, to me, to worry about. Might as well go play the slot machines. I'll have more of a chance of winning.
* DHMO is also known as water. H20. Ya know, that stuff you drink to keep vaping from dehydrating you?
** Airline flight attendants have the highest exposure to radiation, because they're at altitude more than anyone else.