---
title: 'Strange Things Are Happening in Quantum Computing'
source: 'https://youtube.com/watch?v=spRN7DfzhrU'
video_id: 'spRN7DfzhrU'
date: 2026-08-04
duration_sec: 428
---

# Strange Things Are Happening in Quantum Computing

> Source: [Strange Things Are Happening in Quantum Computing](https://youtube.com/watch?v=spRN7DfzhrU)

## Summary

The video critically examines recent developments in quantum computing, highlighting a disconnect between ambitious claims and practical reality. It explains the theoretical basis of quantum advantage, notes the disappearance of specific qubit targets from IBM's roadmap, and critiques the shift towards hybrid approaches and vague AI-related promises. The video also discusses increased government funding and questions the return on investment compared to other technologies like nuclear fusion.

### Key Points

- **Quantum computing basics** [00:27] — Quantum computers use qubits with quantum properties. The exponential growth of states (2^n) is not unique to qubits; the advantage comes from entanglement, allowing superpositions with arbitrary weights, which enables certain calculations much faster.
- **IBM roadmap changes** [01:50] — IBM's earlier roadmap planned over 4,000 qubits by 2025 and scaling to 10,000+ by 2026. Their current roadmap has removed these specific qubit numbers, indicating that plans have not gone as expected.
- **Shift to hybrid approaches** [02:15] — IBM and others have shifted to hybrid approaches combining classical and quantum computers. IBM's recent 'quantum-centric supercomputing' simulation of a protein complex mostly used classical computing, with results comparable to purely classical methods.
- **Vague AI and quantum claims** [03:24] — There are increasingly bizarre claims that AI and quantum computing could deliver personalized medicine or boost AI, which the speaker dismisses as 'sprinkling AI on it' like parsley.
- **Government funding surge** [04:26] — China has included quantum computing in its 5-year plan, and the US recently invested $2 billion into quantum computing, despite the technology having zero practical use so far. Companies like GlobalFoundries and IBM are launching quantum-specific initiatives.
- **Quantum wafers and production** [05:18] — IBM, Google, and Amazon use superconducting circuits as qubits, which can be printed with standard chip production methods. However, chip production is not the main problem; the challenge is using the chips in combination and finding applications.
- **Erosion of use cases** [06:07] — Potential applications like quantum chemistry, material science, logistics, and finance have eroded, either because AI can solve them or because no one can find useful applications even theoretically. The only undisputed use case is breaking old encryption, which is not for the average person.
- **Comparison to nuclear fusion** [06:33] — The speaker argues that billions in quantum computing funding have low expected ROI compared to nuclear fusion, which has a fraction of the funding but clearer ROI, making the quantum investment seem 'crazy'.

### Conclusion

The video concludes that quantum computing is overhyped and underdelivering, with billions in funding despite no practical applications, while more promising technologies like nuclear fusion are underfunded.

## Transcript

Quantum computing is one of the most exciting  new technologies on the radar. And you all know   that I've been following the progress  closely for years. In the past months,   weird things have been happening. And when  I say weird, I don't mean quantum weird,  
where a particle goes through two slits at  once. I mean business weird, where a company   goes through two road maps at once. Let's have  a look. But first, what's a quantum computer?  
Quantum computers perform mathematical operations  with logical units that have quantum properties.   The so-called quantum bits or qubits for short.  I sometimes hear people say this gives quantum  
computers special power because the number of  qubits grows exponentially. If you have a qubit   with two states 0 and 1, then you can combine it  with a second qubit to a product state. Then you  
have 0 0 ,1 1, 0 1, and 0 1. that makes four. Take  a third and you have eight. For n qubits, you have   2 to the n. It indeed grows exponentially.  Yes, but actually that's the case already  
for standard bits. The quantum advantage does not  just come from combining the qubits, but from the   possibility to entangle them. This means not only  do you have 0 0, 1 1, 0 1, and 0 1, you can also  
take sums of those with arbitrary prefactors. say  0 1 + 1 0 with weights to tune. Now you see these   states don't exist in a standard computer. So you  have many many more states you can calculate with  
and this is basically why quantum computers can  perform certain calculations much faster. Alas   this advantage only becomes relevant for a large  enough number of qubits somewhere in the range of  
some 100,000 to a million. One of the diagnostics  I like to look at is IBM's road map to quantum   computing. If you look at their road map from a  few years ago, they were planning that by 2025,  
they'd have more than 4,000 qubits and by 2026,  they'd be scaling to a 10,000 and up. If you look   at their current road map, these qubit numbers  have just disappeared. Clearly, something didn't  
quite go according to plan. Instead, IBM and  others have been quietly shifting goals. First,   they claimed that noisy quantum computers could  have practical uses already. Unfortunately,  
these uses were nowhere to be found. Now, they use  hybrid approaches that combine both conventional   and quantum computers. The advantage of those is  that you can't tell what the quantum part was good  
for. IBM has for example recently announced they  used quantum centric supercomputing to simulate   a big protein complex. Quantum centric is new  speak for hybrid approaches. In this case most of  
the calculation was actually done by a classical  supercomput and they say in the paper themselves   that the results of the purely conventional and  partly quantum computation are comparable. Don't   get me wrong, this is all very interesting and  I think it's cool they're doing this and so on,  
but this is still far off any practical use.  When it comes to practical uses, all we have   are increasingly bizarre claims like that AI and  quantum computing could deliver on the promise of  
personalized medicine or that quantum computers  could boost AI. When in doubt, sprinkle AI on it.   It's like the parsley of business. This video  was made possible by Free Cash. Free Cash is a  
rewards platform where you can earn real money by  completing offers, for example, by testing apps,   answering surveys, or playing games. I tried it  and it's straightforward. You create an account,  
choose an offer, follow the requirements, and  then free cash tracks your progress. They have   several cash out options including PayPal, Visa,  Amazon gift cards, and others. Once you reach the  
minimum payout, you can withdraw your earnings.  And yes, that really works. They're rated 4.7 on   Trustpilot. You're already on your phone all the  time anyway, so free cash is a good and casual way  
to turn that time into something valuable. Try it  out yourself by using my link or scanning the QR   code. Offers and bonuses may vary by country.  And now back to the science news. That doesn't  
deter the governments of the world from pouring  money into it. I already reported recently that   China has made quantum computing part of their  new 5-year plan. Now, rather predictably, the US  
government has ramped up their investments into  quantum computing as well. Just last week, they   put a total of $2 billion into quantum computing,  which I remind you is the technology that so  
far has zero practical use. Global Foundaries  promptly launched Quantum Technology Solutions,   a new quantum business for the quantum industry  and its pipeline of quantum innovators. IBM says  
they're building a quantum foundry for quantum  wafers to power America's quantum leadership,   quantum innovation, and quantum ecosystem.  Wow, so much quantum. Who' not be impressed?  
What the heck is a quantum wafer, you ask? Yes,  good question. IBM and Google and Amazon use   superconducting circuits as qubits. One of their  main benefits is that you can print them using  
standard chip production methods on the standard  silicon wafers. They are not standard microchips.   You need to print the wires with materials that  become superconducting at low temperatures. But   the production of the chips has never been the  main problem. The problem is to use the chips in  
combination and find the use for them. The one  application for large enough quantum computers   that no one really doubts is that they could  break some old encryption protocols. But this is   not exactly a use case for the average person. And  once the old protocols have been decrypted, that's  
that. The other use cases that you often hear,  quantum chemistry, material science, logistics,   and finance have all eroded one after the other.  Either because AI is doing things that quantum  
computing was supposed to solve, or because no  one could actually honestly find something useful   to do with them, even theoretically. To me, all  these billions being thrown at quantum computing  
looks that [ __ ] crazy given the low expected  return on investment. This is particularly obvious   if you compare it to nuclear fusion, which has a  fraction of the funding but dramatically clearer  
return on investment. Yes, I'm now defending  nuclear fusion as the practical option. This   is how weird this has become. Don't forget to  check out free cash using my link or scanning  
the QR code and turn your spare time into real  money. Thanks for watching. See you tomorrow.