I don't know if there are specific laws in place (probably are considering how corrupt everything is) but I would say even though machines come up with the base design it would still need to be approved by a human (or team of humans) who would be liable for any failures that occur due to their AI's design and their negligence for not properly testing it.
I don't know if there are specific laws in place (probably are considering how corrupt everything is) but I would say even though machines come up with the base design it would still need to be approved by a human (or team of humans) who would be liable for any failures that occur due to their AI's design and their negligence for not properly testing it.