---
title: 'Quantum Computing Finally Got Its Numbers, and They’re Bad'
source: 'https://youtube.com/watch?v=t3CvQRH-mFg'
video_id: 't3CvQRH-mFg'
date: 2026-11-11
duration_sec: 398
channel: 'Sabine Hossenfelder'
---

# Quantum Computing Finally Got Its Numbers, and They’re Bad

> Source: [Quantum Computing Finally Got Its Numbers, and They’re Bad](https://youtube.com/watch?v=t3CvQRH-mFg)

## Summary

The video critiques the feasibility of large-scale quantum computing, citing a new paper that calculates the physical resources required for a fault-tolerant quantum computer. The creator argues that the size, cooling requirements, and noise issues make practical quantum computers unlikely, and questions their real-world applications beyond research.

### Key Points

- **US DOE Announces 2028 Quantum Computing Goal** [00:00] — The US Department of Energy, following an executive order, plans to deploy the world's first fault-tolerant, scientifically relevant quantum computer by 2028. The creator sarcastically suggests this is unrealistic.
- **New Paper Calculates Quantum Computer Resource Requirements** [00:41] — A new paper is the first to calculate the physical resources needed for a large error-corrected quantum computer, focusing on superconducting circuits, the most advanced current approach.
- **Quantum Computers Are Mostly Noise Buffers** [01:08] — The visible hardware is mostly cooling equipment for a tiny microchip. Current chips have at most a few thousand qubits, but a serious quantum computer would need millions, making the system enormous.
- **Size Estimate: 800 m² for 4 Million Qubits** [01:50] — Assuming 10,000 qubits per cooling apparatus, a 4-million-qubit quantum computer would cover over 800 m², the size of a German Aldi supermarket, and double that with manipulation and readout equipment.
- **Noise Problem Explodes with Size** [02:41] — The creator argues that error correction becomes impossible at such scales, and alternative approaches like PsiQuantum's photonic qubits have not been demonstrated. Executives of quantum computing firms are selling shares.
- **Wellcome Leap Challenge: No Winner** [03:27] — In 2023, Wellcome Leap offered $5 million for a healthcare problem where a quantum computer beats a conventional one. No one won, indicating a lack of practical use cases.
- **Quantum AI Fails to Beat Classical Algorithms** [04:10] — A Los Alamos team found that conventional algorithms match or beat quantum versions on every benchmark up to 1,000 logical qubits, suggesting quantum AI may not offer advantages.
- **Only Solid Use Case: Cracking Old Ciphers** [04:56] — The creator admits cherry-picking bad news but notes that the only solid use case is breaking old ciphers, which doesn't justify the investment compared to other public benefits.

### Conclusion

The creator remains highly skeptical of quantum computing's practical applications, arguing that the physical and noise challenges are insurmountable and that current investments are misallocated.

## Transcript

Quantum computing has a deadline. Following&nbsp; an executive order of President Trump,&nbsp;&nbsp; the US-American Department of Energy&nbsp; has announced that they will “deploy the&nbsp;&nbsp; world's first fault-tolerant, scientifically&nbsp; relevant quantum computing capability for&nbsp;&nbsp;
research and development by 2028.” Awesome!&nbsp; Can we please also have an executive order&nbsp;&nbsp; for quantum gravity by 2030, and while&nbsp; we’re at it, immortality by 2040? Yeah,&nbsp;&nbsp;
I think it’s not going to happen, and today I&nbsp; want to tell you why. Stay around until the end&nbsp;&nbsp; for a special offer that will bring your coding&nbsp; skills to an entirely new level with boot.dev.
For years I have told you that I think&nbsp; quantum computers are not going to scale&nbsp;&nbsp; and now the authors of a new paper finally put&nbsp; the numbers behind it. It’s the first attempt&nbsp;&nbsp; to calculate the physical resources that&nbsp; a large error-corrected quantum computer&nbsp;&nbsp;
would require. They look specifically&nbsp; at superconducting circuits as qubits,&nbsp;&nbsp; which is the most studied and&nbsp; most advanced current approach. The problem with quantum computers&nbsp; is apparent if you just look at&nbsp;&nbsp;
this footage. This thing which you see&nbsp; here has little to do with computing,&nbsp;&nbsp; it is mostly a noise buffer and equipment&nbsp; to cool a tiny microchip at the very bottom. These chips may contain 1,000 or maybe&nbsp; a few thousand of these qubits at most&nbsp;&nbsp;
before they simply get too large for the&nbsp; cooling apparatus. If you want to build a&nbsp;&nbsp; serious quantum computer that can do&nbsp; all the things we’ve been promised,&nbsp;&nbsp; you’d need a few million qubits. Then the&nbsp; problem becomes you have to combine multiple&nbsp;&nbsp;
of these cooling devices. Leaving aside&nbsp; that no one really knows how to do this&nbsp;&nbsp; without ruining the required entanglement, it&nbsp; becomes a problem simply because of the size.
The authors of the new paper say, let’s&nbsp; look at something like 4 million qubits,&nbsp;&nbsp; which is a reasonable estimate, and be generous&nbsp; and say that we can put about 10,000 qubits in one&nbsp;&nbsp; cooling apparatus. The entire quantum computer&nbsp; would then cover an area of more than 800 m²;&nbsp;&nbsp;
that’s about the size of a German Aldi&nbsp; supermarket. And that is without any of&nbsp;&nbsp; the necessary equipment to manipulate and read&nbsp; out the thing. Add all that equipment on top and&nbsp;&nbsp; we’re talking about at least twice the size, not&nbsp; counting all the extra facilities necessary to&nbsp;&nbsp;
build the components. So if you were thinking we’d&nbsp; have a factory that produces quantum computers&nbsp;&nbsp; like laptops, it’s not going to happen. The&nbsp; factory will be the quantum computer, basically. And with a size like this the noise&nbsp; problem is going to explode. I am sorry,&nbsp;&nbsp;
but really I see no way you will ever&nbsp; error correct anything that size. Now look,&nbsp;&nbsp; there are other approaches to quantum&nbsp; computing that use other technologies,&nbsp;&nbsp; like for example PsiQuantum that wants to use&nbsp; photons as qubits. But no one has even seen a&nbsp;&nbsp;
demo of the thing. In what I am sure are&nbsp; completely unrelated news, top executives&nbsp;&nbsp; of leading publicly traded quantum computing firms&nbsp; have sold off significant numbers of their shares. It is not that I am somehow against quantum&nbsp; computers, I am just extremely sceptical&nbsp;&nbsp;
of their supposed practical uses besides the&nbsp; research opportunity of just studying a big,&nbsp;&nbsp; highly entangled state. And keeping these&nbsp; researchers employed. That, to be fair,&nbsp;&nbsp; is a quantum computing application&nbsp; with excellent experimental support.
Let me give a few timely examples to explain why&nbsp; I am sceptical. In 2023, the American non-profit&nbsp;&nbsp; Wellcome Leap offered $5 millions to anyone&nbsp; who would find a healthcare problem where a&nbsp;&nbsp;
quantum computer could beat a conventional one.&nbsp; They published the results in April. The winner&nbsp;&nbsp; was… no one. No one could find a good use&nbsp; case for a quantum computer in healthcare.
This isn’t an isolated case, I have&nbsp; previously reported that whenever someone&nbsp;&nbsp; looks closely at what a quantum computer&nbsp; can supposedly do the answer is not much. In a recent paper, a team from Los Alamos looked&nbsp; at the question of whether quantum computers are&nbsp;&nbsp;
actually good to speed up neural networks,&nbsp; “quantum AI” as the headlines like to put it.&nbsp;&nbsp; They found that the conventional algorithm&nbsp; would match or beat the quantum version on&nbsp;&nbsp; every benchmark they tried, up to more&nbsp; than1,000logical qubits. So if you want&nbsp;&nbsp;
to make quantum artificial intelligence more&nbsp; efficient, just remove the quantum computer. The authors say that it might be that there&nbsp; are some sorts of datasets for which quantum&nbsp;&nbsp;
computers might have an advantage, but&nbsp; that hasn’t been demonstrated and then&nbsp;&nbsp; there is again the question of who would&nbsp; actually have a use for that. Likewise,&nbsp;&nbsp; salesman problem and related logical and&nbsp; logistic problems or finance applications.
Now look. I openly admit that I am&nbsp; cherry-picking bad news to confirm&nbsp;&nbsp; my biases. But that doesn’t mean this&nbsp; bad news doesn’t exist. The one so far&nbsp;&nbsp; solid use case of quantum computers remains&nbsp; cracking old ciphers. That is not nothing,&nbsp;&nbsp;
but I think it doesn’t remotely justify&nbsp; the investment. It pains me to think what&nbsp;&nbsp; good all this money could have done had it&nbsp; been invested in something that benefits the&nbsp;&nbsp; public. Like science communication that could have&nbsp; prevented this utter insanity in the first place.
Artificial intelligence can do a lot, but&nbsp; if anything, it's increased the demand for&nbsp;&nbsp; software engineers who understand how to&nbsp; code. And there's no better place to get&nbsp;&nbsp; started than with today's sponsor, Boot.dev.&nbsp; Think about everything you've done today,&nbsp;&nbsp;
and chances are most of it was accessing some&nbsp; sort of data through a user interface. Because&nbsp;&nbsp; that's how most apps work, how most businesses&nbsp; operate, and how most research is presented. And&nbsp;&nbsp;
that's exactly what you'll learn on boot.dev,&nbsp; backend development, database management,&nbsp;&nbsp; user interfaces. They have hands-on lessons in&nbsp; which you write code directly in the browser&nbsp;&nbsp;
and you solve challenges to level up. And it's not&nbsp; single scattered lessons. They have very carefully&nbsp;&nbsp; designed learning paths that'll get you started&nbsp; on anything from Python for beginners, Typescript&nbsp;&nbsp;
to your first personal project. If you decide to&nbsp; go for the full experience, use my code SABINE&nbsp;&nbsp; to get 25% off your first year. So go and check&nbsp; this out. Thanks for watching. See you tomorrow.
