The author misunderstands science, math and engineering. He says "Whatever can go wrong, will go wrong," then uses this argument to cite the need to use UBI as "fault tolerance" and cites examples of planes and safety systems.

The irony here is that his statement of whatever "will go wrong" applies to fault tolerant systems as well. You can't deploy redundant safety features without testing that it works.

Is UBI tested to a degree of verifiability that it will always work? No.

Also "testing" and "making arguments" is not strictly the case of engineering. These two things are the domain of science and while engineering can in many cases use "science" this is not what's going on in this article.

Engineering is mostly about using what we already know about in science and mathematics to solve problems.

This article neither uses Engineering, math or science to say anything. It's just a qualitative argument demonstrating an arbitrary point using a bunch of analogies to cast an illusion that this is an "engineering argument."

You need to run a scientific experiment with a control group or observe real world examples to make this argument at all "scientific" or "engineering" related. Until then it's just an opinion.

Whatever your political stance on this policy is... nothing substantial was said here. Just a regurgitation of things we all already know that may or may not be true.