Verified Quantum Advantage: What IBM's Milestone Really Changes
💡 On 30 July 2026, IBM and the University of Chicago achieved verified quantum advantage: a quantum computer using 70 error-corrected logical qubits solved a problem no classical computer can match in reasonable time - and for the first time, they proved the answer was statistically correct. This closes quantum computing's biggest credibility gap.
- 70 error-corrected logical qubits completed 2,415 gate operations in ~15 minutes; classical computers cannot reproduce this in any practical timeframe
- Logical error rates are 10 times lower than physical error rates - error correction is working at meaningful scale
- Previous quantum advantage claims could not prove results were genuine; this one can, with statistical confidence, using the error-correction codes themselves as verifiers
- The problem solved is theoretical, not practical: useful quantum applications need thousands to millions of logical qubits, still years away
- Now is the right time to plan post-quantum cryptography migration - not because the threat is immediate, but because migration takes years
What exactly did IBM and the University of Chicago just demonstrate?
The paper is titled "Sampling hard circuits with verifiably high fidelity." That title contains the entire story: a quantum circuit that is simultaneously hard to simulate classically and verifiable as genuinely quantum - closing the gap that had made verified quantum advantage a contested idea.
The team encoded 70 logical qubits from many physical qubits and ran 2,415 logical two-qubit gate operations plus 468 "T gates" - a standard measure of circuit complexity. The computation finished in approximately 15 minutes. A classical supercomputer cannot reproduce that result in any practical timeframe.
The logical error rates reached 10 times lower than the physical error rates of the underlying hardware. Error correction is suppressing noise, not amplifying it - at a scale that is now measurable and public.
Why is verification such a big deal?
In 2019, Google claimed quantum advantage with its Sycamore processor. Critics immediately identified a flaw: the method used - random circuit sampling - produces outputs that look random whether the quantum computer is working correctly or not. You could not tell genuine quantum results from well-calibrated noise.
IBM's structured approach breaks this deadlock. By building internal structure into the circuit design, the researchers made the error-correction codes themselves reveal whether the computation ran faithfully. The circuit is both hard to fake and self-auditing.
Bill Fefferman, the University of Chicago professor who led the theory side: "Verification remains one of the biggest challenges in firmly establishing experimental quantum advantage." IBM's quantum chief Jay Gambetta stated: "We are now firmly in the quantum advantage era. We have demonstrated a quantum computation beyond the practical reach of classical computers that establishes, with statistical confidence, a lower bound on how faithfully it was executed."
What does this mean for encryption - should you be concerned?
The honest answer: not yet, but start planning.
The RSA and elliptic-curve cryptography that secures HTTPS, banking, and email relies on mathematical problems that classical computers cannot solve efficiently. A large enough quantum computer could solve some of these. The key phrase is "large enough": breaking RSA-2048 is estimated to require roughly 4,000 error-corrected logical qubits running millions of operations - orders of magnitude beyond today's 70.
What has genuinely changed: the trajectory of error correction now looks more credible. IBM's roadmap targets thousands of logical qubits over the coming years. The US National Institute of Standards and Technology finalized post-quantum cryptography standards in 2024 - CRYSTALS-Kyber, CRYSTALS-Dilithium, and SPHINCS+. Organizations handling sensitive long-lived data should begin migrating now.
What quantum computers cannot do yet - the honest limits
This demonstration solved a problem designed to be hard to classically simulate - not a problem with a useful real-world answer. The 70-logical-qubit result is foundational science, not a deployed application.
Useful quantum computing at scale requires:
- Thousands to potentially millions of error-corrected logical qubits for real chemistry or optimization problems
- Even lower error rates for longer, deeper circuits
- Classical-quantum software that translates practical problems into circuits efficiently
IBM's result is foundational, but quantum computing remains years from displacing classical computers for most tasks. This announcement represents the end of the beginning, not the beginning of the end. Each major quantum milestone has arrived roughly on schedule - but always with important caveats that press releases tend to omit.
What should you watch or do next?
IBM published the circuits and results openly via the Quantum Advantage Tracker. The next milestone to watch: independent researchers verifying the claim in the coming weeks. If the verification holds, this will stand as a genuine landmark - the first quantum advantage demonstration that answers its own critics by design.
For anyone managing organizational data security: start your post-quantum cryptography inventory now. Map which systems rely on RSA or elliptic-curve. NIST's standards are ready. Migration takes years, and the window to start comfortably is open today.
FAQ
Will quantum computers break my passwords soon?
Not soon. Breaking standard password hashing or common encryption needs millions of error-corrected logical qubits, far beyond today's 70. The risk is real but years - likely decades - from threatening everyday passwords. Long-lived secrets and government data are higher-priority concerns.
How is this different from Google's 2019 quantum advantage claim?
Google's Sycamore used random circuit sampling, producing outputs indistinguishable from noise without internal verification. IBM's structured approach makes the circuit self-verifying: the error-correction codes confirm the computation ran correctly, a dual proof Google's method lacked.
How many logical qubits does a useful quantum computer actually need?
For breaking RSA-2048: ~4,000 logical qubits with millions of clean gate operations. For simulating useful molecules in drug discovery: estimates range from tens of thousands to millions. IBM demonstrated 70 - meaningful as a proof of principle, not yet at application scale.
Are banks and governments already responding to quantum risk?
Yes. The US, EU, and China are all funding post-quantum cryptography migration programs. NIST finalized quantum-resistant standards in 2024. Major financial institutions are inventorying quantum-vulnerable systems. Preparation is real but unevenly distributed across sectors.
What is the Quantum Advantage Tracker?
It is an open platform where IBM publishes circuits, results, and methodology from quantum advantage demonstrations. Independent researchers can download and attempt to replicate or challenge the results. Transparency is part of what makes this claim different from previous ones.
Source(s): IBM Newsroom (2026); The Quantum Insider (2026)
About the author
Dao Huy (Lucas) is a professional translator working across English, Vietnamese, Chinese, and French, with over seven years of experience in technical, legal, and IP translation. He follows developments at the frontier of computing and science because the tools that change how knowledge is created and shared directly shape what translation and localization need to do. When a quantum breakthrough rewrites the rules of secure communication, the documents that govern that transition - standards, specifications, legal frameworks - need precise multilingual expression.
If you need English-Vietnamese or technical translation for software documentation, patents, or IP filings, Lucas offers professional translation services at daohuy.com. Reach out for a quote.
Written by Dao Huy (Lucas), Vietnamese translator & localization specialist (EN · ZH · FR → Vietnamese). See translation services →