It’s not that simple. Imagine you’re dying of a rare terminal disease. A pharma company is developing a new drug for it. Obviously you want it. But they tell you you can’t have it because “we’re not releasing it until we know it’s good”.
As an exercise to remove the bias from this, replace self driving cars with airbags. In some rare cases they might go off accidentally and do harm that wouldn’t have occurred in their absence. But all cars have airbags. More and more with every generation. If you are so cautious about accidental detonations that you choose not to install them in your car, then you’re being too cautious.
I can’t agree that they’re not being cautious enough. I didn’t even read the article. I’m just arguing about the principle. And I don’t have a clue what the right penalty would be. I would need to be an actuary with access to lots of data I don’t have to figure out the right number to provide the right deterrent.
There’s actually a backfire effect here. It could make companies too cautious in rolling out self driving. The status quo is people driving poorly. If you delay the roll out of self driving beyond the point when it’s better than people, then more people will die.
Actually we have “right to try” laws for the scenario I described.
But the FDA could use some serious reform. Under the system we have, an FDA approval lumps together the determinations of whether a drugs is safe, effective and worth paying for. A more libertarian system would let people spend their own money on drugs that are safe even if the FDA’s particular research didn’t find them effective. And it wouldn’t waste tax payer money on drugs that are effective but exorbitantly expensive relative to their minimal effectiveness. But if a wealthy person wants to spend their own money, thereby subsidizing pharmaceuticals for the rest of us, that’s great in my opinion.