How IBM will build the world's first large-scale fault-tolerant quantum computer

[John Clark]

IBM just published a roadmap of what they expect to accomplish in the next few years in the field of quantum computers, and it's pretty damn ambitious: 

How IBM will build the world's first large-scale, fault-tolerant quantum computer

After reading the above I decided to ask Claude the following question: 

"IBM recently predicted that by 2029 they should have a quantum computer that has 200 error corrected logical Qubits, and by 2033 they will have 2000. What sort of things could something like that do, could it break RSA encryption?"

Claude responded with the following: 

"IBM's quantum computing roadmap is indeed ambitious. Starling is expected to be capable of 100 million quantum operations using 200 logical qubits 'The science is solved': IBM to build monster 10,000-qubit quantum computer by 2029 | Live Science, with the larger Blue Jay system planned for 2033 featuring 2,000 logical qubits capable of 1 billion quantum operations.

These systems could potentially tackle several important computational problems:

What 200-2000 logical qubits could accomplish:

Chemistry and Materials Science: Simulating molecular interactions for drug discovery, catalyst design, and materials development. This could accelerate pharmaceutical research and help design better batteries, solar cells, and superconductors.

Optimization Problems: Solving complex logistics, supply chain optimization, financial portfolio management, and traffic routing problems that are computationally intensive for classical computers.

Machine Learning: Quantum machine learning algorithms could potentially offer advantages for certain pattern recognition and data analysis tasks.

Cryptanalysis: This brings us to your RSA question - and it's a crucial one.

Regarding RSA Encryption:

The threat to RSA depends on successfully implementing Shor's algorithm at scale. Current estimates suggest that breaking RSA-2048 (the current standard) would require approximately 2,000-4,000 logical qubits with millions of quantum operations.

IBM's 2029 system with 200 logical qubits would likely be insufficient to break current RSA encryption. However, their 2033 system with 2,000 logical qubits could potentially approach the threshold needed for RSA-2048, depending on error rates and algorithm efficiency improvements.

This timeline aligns with why cryptographers are already developing "post-quantum" encryption methods that would resist quantum attacks. The transition needs to happen before quantum computers become capable enough, not after.

It's worth noting that these are IBM's projections, and quantum computing development faces significant technical challenges in error correction and maintaining quantum coherence at scale."

===

2025 IBM Quantum Roadmap update

Realizing large-scale, fault-tolerant quantum computing

 John K Clark    See what's on my new list at  Extropolis

3a9

[Brent Meeker]

I would expect that government spy organizations would already be archiving the RSA encrypted data traffic in anticipation of being able to read it in 4 to 8 years.

Brent

--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/everything-list/CAJPayv2N0wojZ2x2vykvaDFMuPw7vYsJ5jj4a%2Bjt9OAaoq3eqw%40mail.gmail.com.

[John Clark]

On Tue, Jun 10, 2025 at 9:20 PM Brent Allsop <brent....@gmail.com> wrote:

> Interesting. So bitcoin doesn't need to worry too much about that for now.  And even when/if it does need to start worrying about it, as it pointed out, solving cryptography is much easier than the fact that "quantum computing development faces significant technical challenges in error correction and maintaining quantum coherence at scale."

It's not certain but Bitcoin might run into trouble well before 2033. At the moment IBM seems to be leading in the race to build a Quantum Computer and, because of the technology it's using, its rate of progress is relatively easy to predict, but that's not true for one of IBM's competitors.  Microsoft is using a technology that is radically different from IBM's, they're trying to use non-Abelian Majorana Fermions to make a Topological Quantum Computer. Microsoft made a risky bet about 10 years ago and I admire their guts, their unusual approach may prove to be impractical, but if it's workable then they would be able to make progress toward a Quantum Computer far faster than IBM, or anybody else.

It takes time so If the Bitcoin people had been smart and cautious they would have changed their cryptography to something more secure several years ago, but they haven't even started.  

 John K Clark    See what's on my new list at  Extropolis

5g2