Quantum Computers

Once these take off this is when the AI is for real going to take over, right?

Attached: d41586-021-03476-5_19875844[1].jpg (800x533, 178.25K)

>Once these take off

Attached: 1614863705191.jpg (696x535, 65.04K)

those can barely "do" something

that image is fake

>those can barely "do" something

Attached: bell-labs-first-transistor-640x573[1].jpg (640x573, 44.29K)

lol, lmao

j-just two more years!

>compare this to a bunch of golden wires in a giant fridge

Quantum computations are not more efficient for neural nets. It's important to understand that quantum computers are more like a kind of domain-specific processor than a "faster conventional computer"

Theories are not proven, nice rube goldberg machine OP, looks like a bunch of gold plated SMA connectors going to nothing. To date, no quantum computer has computed a single thing. Quantum computing languages are emulated on x86 and have no purpose, nor does the money for "quantum theory research"

Attached: 1627358458533.gif (502x400, 2.49M)

prove it

Quantum computers don't mean "faster, and better ai", they mean "native/more efficient computation of quantum algorithms."

>Problems which are undecidable using classical computers remain undecidable using quantum computers.[4]:127 What makes quantum algorithms interesting is that they might be able to solve some problems faster than classical algorithms because the quantum superposition and quantum entanglement that quantum algorithms exploit probably cannot be efficiently simulated on classical computers (see Quantum supremacy).
>might be able to solve some problems faster than classical algorithms because the quantum superposition and quantum entanglement that quantum algorithms exploit probably cannot be efficiently simulated on classical computers (see Quantum supremacy).
>might
>probably

It's just specific algos that maybe will be faster.

The real question is, can it run Minecraft and simulate a Minecraft GF for me

Doubt. Legitimate AI could unironically come to be if we somehow go back to analog computers and bring them to (or past) the level of high-end digital computers. Our current methods are ultimately a dead-end, don't buy into the lies/speculations of birthing AI through quantum computing. Anything that will be advertised as one will guaranteed to be marketing bs.

aren't the algorithms pretty important though? like simulations and code breaking?

i absolutely hate people who say this. why is it an unfair assumption? just because u dont like it? are u retarded? i would spit in ur direction if i could

bamp

No ai (matrix multiplication) will be sped up analogue chips, like repurposed flash storage. This technique can already get the same speeds as top of the line graphics cards with, again, dirt cheap flash storage.

What's amazing is, I can run qbit quantum code on my PC. Multiple cores can accomplish much of the same functions as qbits. The more we can cluster together the more we can accomplish. And I don't need a hydrogen pump and 9kw to run it.

What's more is gpu ai is pulling ahead in terms of solving these equations faster than a 40 qbit CPU.

Yeah it can solve scientific solutions not casual problems.

quantum branch prediction is going to be wild

Exactly, which is exactly why the glowies of the world get a little stiff at the thought of having it first.