The testing here is not just that it is safe on skin, but the SPF test itself is done by slathering it on humans and exposing them to light to determine a rating.

It's not side-effects that are in question here, it's the intended effect. When it comes to its effectiveness at blocking UV, there should be a better way than just "apply some to a dozen random volunteers and time how long it takes before they get a sunburn".

In my imagination, the lab would have some a testing process that spreads a precisely-controlled volume over a standard surface area, textured to be similar to skin, then measures UV transmission percentage vs wavelength with a diffraction grating and photocell. Or something like that!

I've actually been a test subject for sunscreen here in Australia. It involved having sunscreen put on different parts of your body, hopping into a Jacuzzi for an hour or so, then being zapped with UV on both sunscreened and clear skin.

Easiest hundred bucks I ever made, gotta say.

You're kidding? You equate "not testing on human subject" to "not testing"?

However across millions of people over lifetimes may offer substantial increase in incident reduction no?

what a tremendous failure on a regulation level

As others have mentioned, the difference between 75% (SPF 4) and 96% (SPF 30) might seem small, but the latter implies you can stay in the sun 7.5 times as long before getting sunburnt. That's significant. Moreover sunscreen rapidly loses effectiveness, so having "extra" protection might be worth it, especially if you don't reapply every 2 hours or after sweating/swimming, which what most sunscreens recommend.

I mean, that is what i thought it meant.

So SPF 4 you are letting 25% of the sun through. I would assume that would be enough to still be sun burnt on a high uv index day if you spend most of it on the beach.

What is of interest is not the blocking percentage, but the transmission percentage.

According to WikiPedia:

"For example, "SPF 15" means that 1⁄15 of the burning radiation will reach the skin, assuming sunscreen is applied evenly at a thick dosage of 2 milligrams per square centimeter[67] (mg/cm2)."

so assuming a linear dose response relationship (obviously oversimplified) when not using the sunscreen 15 times more instantaneous random damage is incurred compared to when using the sunscreen.

This does not translate directly into the rate of cancers though: just like the final damage of a meteorite storm isn't proportional, even though the instantaneous damage is.

Suppose a meteorite strikes a hospital, lots of damage. Then years later a meteorite strikes a school, lots of damage. Obviously if both happen in quick succession more damage will occur.

But if the whole human population takes up sunscreen use, selective pressure on cellular coping mechanisms will be relaxed, and eventually future generations won't be as resilient against sunburn. So just live your life, and don't allow scaremongers to separate you from your money, or thus indirectly scare you into doing your job for them.

But what about the sunscreen with ingredients that are carcinogenic before you even need to consider UV protection?

You can't compare the straight percentage, a 98% filter lets through twice as much as a 99% filter.

According to the chart it's not even that bad? Sure, it underperformed its claimed SPF, but it's still above median.

I wear a baseball cap and a hooded sun shirt. Plus mineral sunscreen all over my face and neck.

https://hendersonvilleoutfitters.com/products/upf-shirts