Neuralink is at best replicating decades-old research, all while horrifically violating scientific codes of ethics,

Boring is just a tunnel digging company (tunnel-boring is a very often unavoidable construction technique and thus is a very mature technology, he's not going to magically find an order-of-magnitude technical breakthrough) whose basic premise is that you can increase throughput per dollar by digging lots of slightly-smaller tunnels for cars instead of a single subway-sized car; a premise that is objectively false.

Solarcity might have been an interesting idea but was handled poorly and went bankrupt, at which point Musk forced Tesla to buy it so he personally could avoid the bad PR of "Musk's company goes bankrupt!", which is quite illegal and an abuse of his power as CEO of Tesla.

Starlink flat-out makes a loss per satellite launched, and scaling its userbase up won't fix that; it's dead once it runs out of investor money.

There's this weird situation were Musk has N different ventures, and each venture looks alright on the surface but is rotten when you take a closer look - but everyone thinks "well sure, but the other (N-1) ventures are doing fine, so maybe I'm missing something about this one?", not realizing that all the other ventures are just the same.

Please, please do the following exercise: ignore all the PR you've heard about Musk, and pick one of his companies that's not Tesla and look closely at it. Evaluate if that company makes sense on its own merits, without Musk.