On the part about computations not running in parallel universes, this is precisely what was suggested in the MIT Press:
"How could such a computer carry out calculations?…Deutsch’s answer is that the calculation is carried out simultaneously on identical computers in each of the parallel universes corresponding to the superpositions. For a three-qbit computer, that means eight superpositions of computer scientists working on the same problem using identical computers to get an answer. It is no surprise that they should “collaborate” in this way, since the experimenters are identical, with identical reasons for tackling the same problem…It is a matter of choice whether you think that is too great a load of metaphysical baggage. But if you do, you will need some other way to explain why quantum computers work."
Of course, the MIT Press didn’t assert this claim with absolute certainty, but it did leave the door open.
I don’t know enough to comment but I think there is a good discussion to be had here. What are we missing?
"Under the many worlds interpretation (which is subscribed to by about 50% of quantum foundations experts) the process of splitting into different universes only happens during measurement or when a quantum system interacts a lot with its environment, leading to decoherence. Both of those are precisely NOT the things you want to happen during quantum computation.
Adding to the confusion, some many worlds physicists prefer not use the word "measurement", they only talk about decoherence. Furthermore, the splitting into separate universes in many worlds not a binary thing, it's more fuzzy, and at very short time scales interactions between the different branches is possible. In any case, this splitting into different universes is not what happens in quantum computation. It's just dead wrong, but this is often repeated in the popular presses.
I talked to GPT-4o about this, and it points out that you could interpret quantum computation as involving different branches ("universes") interacting a lot. But in my opinion, it doesn't make sense to talk about "separate universes" if they are interacting heavily. The MIT press article conjures up a picture of identical computers in independent universes each doing different computations. That's wrong."
Sabine Hossenfelder YT video is incorrectly linked
fixed, thanks!
On the part about computations not running in parallel universes, this is precisely what was suggested in the MIT Press:
"How could such a computer carry out calculations?…Deutsch’s answer is that the calculation is carried out simultaneously on identical computers in each of the parallel universes corresponding to the superpositions. For a three-qbit computer, that means eight superpositions of computer scientists working on the same problem using identical computers to get an answer. It is no surprise that they should “collaborate” in this way, since the experimenters are identical, with identical reasons for tackling the same problem…It is a matter of choice whether you think that is too great a load of metaphysical baggage. But if you do, you will need some other way to explain why quantum computers work."
Of course, the MIT Press didn’t assert this claim with absolute certainty, but it did leave the door open.
I don’t know enough to comment but I think there is a good discussion to be had here. What are we missing?
I responded to you on X in a short series of tweets (https://x.com/moreisdifferent/status/1879161951508680965), but I'll duplicate my comment here for others:
"Under the many worlds interpretation (which is subscribed to by about 50% of quantum foundations experts) the process of splitting into different universes only happens during measurement or when a quantum system interacts a lot with its environment, leading to decoherence. Both of those are precisely NOT the things you want to happen during quantum computation.
Adding to the confusion, some many worlds physicists prefer not use the word "measurement", they only talk about decoherence. Furthermore, the splitting into separate universes in many worlds not a binary thing, it's more fuzzy, and at very short time scales interactions between the different branches is possible. In any case, this splitting into different universes is not what happens in quantum computation. It's just dead wrong, but this is often repeated in the popular presses.
I talked to GPT-4o about this, and it points out that you could interpret quantum computation as involving different branches ("universes") interacting a lot. But in my opinion, it doesn't make sense to talk about "separate universes" if they are interacting heavily. The MIT press article conjures up a picture of identical computers in independent universes each doing different computations. That's wrong."