Quantum computing has spent years being described as “the technology of the future.” That future is starting to look a little less distant.
The latest quantum computing breakthroughs are no longer limited to experiments that simply prove quantum computers can do something unusual. Researchers are increasingly focused on the harder questions: Can quantum computers correct their own errors? Can they scale to useful numbers of qubits? And can they solve meaningful problems that classical machines cannot handle efficiently?
Some of the most important progress came in 2024, particularly with Google’s Willow processor, IBM’s Heron systems, advances in quantum error correction, and the arrival of standardized post-quantum cryptography. Since then, research has continued into logical qubits, quantum advantage, photonic systems, topological qubits, and new approaches to scaling quantum hardware.
So, what actually happened—and what does it mean?
Let’s break down the biggest developments without getting buried under quantum jargon.
What Are the Latest Quantum Computing Breakthroughs?
Before looking at individual discoveries, it helps to understand what scientists are actually trying to accomplish.
A conventional computer uses bits that represent either 0 or 1. A quantum computer uses qubits, which can exploit quantum effects such as superposition and entanglement.
That sounds incredibly powerful, but there is a catch.
Qubits are extremely sensitive to noise and their environment. A useful quantum computer therefore needs much more than a large qubit count. It needs high-quality qubits, accurate operations, fast control, reliable measurements, and—most importantly—effective quantum error correction.
That’s why many of the most significant recent breakthroughs aren’t simply about adding more qubits.
They’re about making those qubits more reliable.
Google Willow and the 2024 Quantum Computing Breakthrough
One of the biggest headlines in 2024 came from Google Quantum AI.
In December 2024, Google introduced Willow, a 105-qubit superconducting quantum processor. The company reported two major achievements: improved quantum error correction as the system scaled and an extremely fast result on a specialized benchmark known as random circuit sampling.
The error-correction result is arguably the more important of the two.
Google demonstrated that as its encoded quantum system became larger—from 3×3 to 5×5 and then 7×7 arrangements—the logical error rate decreased rather than increasing. Google reported roughly a factor-of-two reduction in error rate at each increase in code size.
This is known as operating below the error-correction threshold.
Why Is “Below Threshold” So Important?
Imagine trying to build a bridge from thousands of tiny pieces, where every piece has a small chance of failing.
Adding more pieces would normally create more opportunities for failure.
Quantum computers face a similar problem. Adding physical qubits can introduce more sources of noise, errors, and control problems.
Quantum error correction is supposed to reverse that relationship.
Once a system operates below the threshold, adding physical resources can make the encoded or logical qubit increasingly reliable.
That’s the breakthrough researchers have been pursuing for decades.
Nature described Google’s 2024 result as a key milestone because it demonstrated that error-corrected quantum calculations could become more accurate as the system grew.
Willow’s Five-Minute Benchmark
The other headline-grabbing Willow result involved a specialized benchmark.
Google reported that Willow completed a random circuit sampling calculation in less than five minutes that, according to the company’s comparison, would take one of today’s fastest classical supercomputers around 10 septillion years.
That number is enormous, but there’s an important qualification.
Random circuit sampling is designed specifically as a difficult benchmark for classical simulation. It is not the same thing as demonstrating that a quantum computer can perform a useful medical, financial, engineering, or scientific calculation faster than a conventional computer.
In other words, it was an impressive demonstration of quantum computational performance—but it wasn’t the end goal.
The real prize is useful quantum advantage.
IBM Heron and the Push Toward Better Quantum Performance
Google wasn’t the only company making significant progress in 2024.
IBM continued improving its superconducting quantum processors through its Heron family.
At its 2024 Quantum Developer Conference, IBM reported that its second-generation Heron processor had 156 qubits and could accurately run quantum circuits containing up to 5,000 two-qubit gate operations. This was part of IBM’s previously announced 100×100 performance challenge.
That matters because qubit count alone doesn’t tell you how useful a quantum processor is.
A machine might have hundreds of qubits, but if errors accumulate after only a small number of operations, those qubits aren’t capable of running long, meaningful algorithms.
IBM’s progress therefore highlights another major theme in the latest breakthroughs in quantum computing 2024:
The industry is moving from counting qubits toward measuring what those qubits can actually accomplish.
IBM’s 2024 research review also reported improvements in circuit execution speed, error rates, modular coupling technologies, and the Qiskit software ecosystem.
Quantum Error Correction Is Becoming the Main Story
For years, quantum computing headlines often focused on one metric:
How many qubits does the machine have?
That’s changing.
Today, researchers increasingly care about logical qubits.
A physical qubit is an individual hardware element. A logical qubit uses multiple physical qubits and error-correction techniques to represent quantum information more reliably.
The goal is not necessarily to have the biggest possible collection of physical qubits.
The goal is to create enough high-quality physical qubits to construct logical qubits that can survive long computations.
Physical Qubits vs. Logical Qubits
Think of it this way:
- Physical qubit: One individual quantum computing element.
- Logical qubit: Quantum information protected using multiple physical qubits and error-correction procedures.
- Fault-tolerant quantum computer: A system capable of carrying out long computations while controlling errors sufficiently well to remain reliable.
This is why Google’s Willow result attracted so much attention.
The important achievement wasn’t simply “105 qubits.”
It was showing evidence that increasing the size of the error-correcting code could actually improve logical performance.
AI Is Also Helping Quantum Error Correction
Another interesting development from 2024 was the use of artificial intelligence in quantum error correction.
Google DeepMind and Google Quantum AI introduced AlphaQubit, an AI-based decoder designed to identify errors in quantum computations.
Google reported that AlphaQubit produced fewer errors than several existing decoding approaches in its tests.
Why does that matter?
Quantum computers generate complicated error patterns. Detecting and interpreting those patterns quickly is itself a computational problem.
Machine learning could potentially help identify which errors occurred and determine how the system should respond.
This doesn’t mean AI has “solved” quantum error correction. It hasn’t.
But it demonstrates an increasingly important relationship between two major technologies: artificial intelligence and quantum computing.
Latest Breakthroughs in Quantum Computing 2024: The Security Side
Not every quantum breakthrough involved building a quantum processor.
One of the most practical developments of 2024 happened in cybersecurity.
In August 2024, the U.S. National Institute of Standards and Technology finalized three post-quantum cryptography standards: FIPS 203, FIPS 204, and FIPS 205.
These standards are designed to protect digital information against future attacks from sufficiently powerful quantum computers.
Why is that necessary?
Some of today’s public-key cryptographic systems rely on mathematical problems that a sufficiently capable quantum computer could potentially solve much more efficiently.
That doesn’t mean today’s passwords suddenly became useless because of Willow or another current processor.
The concern is longer term.
Organizations need time to identify vulnerable systems, replace cryptographic components, test new algorithms, and update infrastructure.
NIST therefore recommends beginning the transition to post-quantum cryptography rather than waiting until a cryptographically relevant quantum computer exists.
Google’s Quantum Echoes and Verifiable Quantum Advantage
The story didn’t stop with Willow.
In October 2025, Google announced another major result involving the Willow processor: an algorithm called Quantum Echoes.
Google reported that the algorithm achieved what it described as the first verifiable quantum advantage on hardware, running 13,000 times faster than a classical algorithm used for comparison on one of the world’s fastest supercomputers.
The important word here is verifiable.
Quantum advantage claims have to answer a difficult question:
How do we know the quantum computer actually produced the correct result?
Google’s Quantum Echoes work was designed around that issue and involved an out-of-order time correlator algorithm.
The potential scientific applications are particularly interesting because the technique could help researchers study the structure and behavior of physical systems, including molecules and other quantum systems.
That’s closer to the kind of problem researchers ultimately want quantum computers to solve.
IBM and Logical Quantum Computing in 2026
By 2026, the emphasis on logical quantum computing had become even stronger.
In July 2026, IBM and researchers at the University of Chicago announced a demonstration involving 70 logical qubits and a computation that the researchers said was beyond the practical reach of leading classical simulation methods. The reported computation took approximately 15 minutes.
The significance wasn’t simply the number 70.
The researchers emphasized two requirements for quantum advantage:
- The computation needs to go beyond practical classical simulation.
- There needs to be evidence that the quantum result can be trusted.
That second point is crucial.
A quantum computer producing an answer that nobody can verify isn’t particularly useful for science or industry.
Reliable quantum computation requires both computational power and confidence in the result.
Microsoft’s Majorana Quantum Computing Approach
Another major development has involved an entirely different approach to building qubits.
In February 2025, Microsoft announced Majorana 1, a quantum processor based on what it called a Topological Core architecture. The company argued that its approach could eventually support quantum systems with up to a million qubits.
The basic idea is fascinating.
Instead of trying to protect fragile quantum information primarily through conventional error correction, topological quantum computing aims to encode information in a way that could make it inherently more resistant to certain errors.
But there’s an important caveat.
Microsoft’s claims have faced skepticism from researchers.
Nature reported in 2025 that some physicists questioned whether the evidence established the existence of the claimed topological qubits.
The debate has continued.
In 2026, Microsoft unveiled an upgraded Majorana 2 chip, but Nature reported that researchers remained skeptical about the company’s broader topological-qubit claims.
That doesn’t mean the underlying research is meaningless.
It means the scientific distinction between an exciting experimental result and a demonstrated scalable technology is extremely important.
A Different Direction: Photonic Quantum Computing
Superconducting qubits aren’t the only path forward.
Photonic quantum computing uses particles of light—photons—to carry quantum information.
This approach has attracted considerable attention because photons can travel long distances and potentially work with existing optical technologies.
Research published in 2025 included advances toward manufacturable photonic quantum-computing platforms, showing that researchers are actively exploring ways to build quantum hardware using semiconductor-style manufacturing approaches.
The attraction is obvious: if quantum processors can eventually be manufactured using scalable processes, it could change the economics of building large machines.
But, again, manufacturing scalability and computational scalability are two different problems.
A technology can be easy to manufacture and still face difficult challenges in error correction, control, networking, and algorithm execution.
Quantum Computers Are Exploring New States of Matter
Some of the most interesting quantum breakthroughs aren’t about traditional algorithms at all.
In 2025, researchers demonstrated a Floquet topologically ordered state on a superconducting quantum processor.
The experiment involved a periodically driven system and allowed researchers to observe phenomena including chiral edge modes and emergent anyonic behavior. The experiment was performed on quantum processors at scales where classical simulation becomes increasingly challenging.
This is important because quantum processors aren’t merely potential future computers.
They can also act as experimental laboratories.
Researchers can use them to create and investigate quantum states that are difficult to study or simulate using ordinary computers.
That opens another possible role for quantum computing: quantum simulation.
What Do These Breakthroughs Actually Mean?
It’s easy to read a headline about quantum advantage and conclude that quantum computers are about to replace classical computers.
That’s not what’s happening.
The latest quantum computing breakthroughs suggest something more nuanced.
The field is gradually moving through several stages:
Stage 1: Demonstrating Quantum Behavior
Researchers first had to prove that quantum processors could perform calculations that classical systems struggle to reproduce.
Stage 2: Improving Hardware
The next challenge was increasing qubit quality, connectivity, coherence, gate fidelity, and control.
Stage 3: Error Correction
Researchers are now demonstrating increasingly convincing methods for turning unreliable physical qubits into more reliable logical qubits.
Stage 4: Useful Quantum Advantage
The ultimate goal is to run algorithms that solve valuable problems better, faster, or more efficiently than classical alternatives.
We’re seeing early progress toward this stage, but the journey isn’t finished.
Where Could Quantum Computing Make the Biggest Difference?
The potential applications are broad.
Drug Discovery
Quantum computers could eventually help researchers model molecular interactions and chemical systems that become difficult to simulate classically.
Materials Science
Quantum simulation could help scientists investigate new materials for batteries, electronics, energy systems, and other technologies.
Chemistry
Because molecules themselves obey quantum mechanics, quantum computers are naturally suited to certain types of molecular simulation.
Optimization
Some complex scheduling, logistics, and resource-allocation problems may eventually benefit from quantum algorithms.
Cryptography
Quantum computing could threaten some existing cryptographic systems, which is why post-quantum cryptography is already being standardized.
Fundamental Physics
Quantum processors can also be used to explore quantum phenomena that are difficult to reproduce with classical machines.
But there’s a key word to remember:
Eventually.
Most of these applications still require substantially more capable and reliable quantum computers than those generally available today.
Why Quantum Computing Is Still So Difficult
Despite all the headlines, major obstacles remain.
1. Qubit Errors
Quantum states are fragile. Noise can destroy useful information.
2. Scaling
Building one good quantum processor is difficult. Building a very large system with consistently high-quality components is much harder.
3. Error-Correction Overhead
Logical qubits can require many physical qubits and substantial control infrastructure.
4. Software
Quantum algorithms are fundamentally different from ordinary software, and programmers still need new techniques and tools.
5. Classical Infrastructure
Quantum computers don’t operate alone. They depend on classical electronics, control systems, cooling, measurement, and data-processing infrastructure.
6. Proving Advantage
It’s not enough to claim that a quantum machine is faster.
Researchers need convincing comparisons and, ideally, ways to verify that the quantum computation produced the correct answer.
What Should We Watch Next?
If you want to follow the latest quantum computing breakthroughs, don’t focus only on headline qubit counts.
Watch these measurements instead:
- Logical qubit quality
- Logical error rates
- Error-correction thresholds
- Circuit depth
- Gate fidelity
- Useful algorithm performance
- Quantum-classical comparisons
- Verification methods
- Manufacturing scalability
- Interconnect technology
These metrics tell you much more about whether quantum computing is moving toward practical use.
Frequently Asked Questions
What is the biggest quantum computing breakthrough in 2024?
One of the most important was Google’s Willow demonstration of below-threshold quantum error correction. The company showed that increasing the size of its error-correcting code reduced the logical error rate, a major milestone toward fault-tolerant quantum computing.
What were the latest breakthroughs in quantum computing 2024?
Major developments included Google’s Willow processor and its error-correction results, IBM’s 156-qubit Heron processor and 5,000-gate circuit demonstration, AI-assisted quantum error decoding, and NIST’s finalized post-quantum cryptography standards.
Has quantum computing achieved useful quantum advantage?
There have been increasingly significant demonstrations of quantum advantage, but the phrase needs careful definition. Google’s 2025 Quantum Echoes experiment was presented as a verifiable quantum advantage for a specialized algorithm, while IBM and University of Chicago reported a 2026 demonstration involving logical qubits and a classically intractable problem. These are important milestones, but they do not mean quantum computers have broadly surpassed classical computers for everyday tasks.
What is a logical qubit?
A logical qubit is quantum information protected by encoding it across multiple physical qubits and using error-correction techniques. Logical qubits are important because useful quantum computers will need to perform long calculations without errors accumulating faster than they can be corrected.
Why is quantum error correction so important?
Physical qubits are highly sensitive to noise. Error correction provides a way to detect and correct certain errors while preserving the information needed for a calculation. Google’s Willow result was significant because its experiments demonstrated improved logical performance as the error-correcting code became larger.
Is Google’s Willow quantum computer better than a supercomputer?
For the specific random circuit sampling benchmark Google tested, Willow produced a result in under five minutes that Google estimated would take a leading classical supercomputer an extraordinarily long time. However, that does not mean Willow is faster than a conventional computer for general-purpose computing.
What is Microsoft’s Majorana 1?
Majorana 1 is Microsoft’s quantum chip based on its proposed topological quantum-computing architecture. Microsoft says the approach could eventually support very large quantum systems. However, researchers have raised questions about the evidence supporting Microsoft’s topological-qubit claims, so it should be regarded as an active and debated research direction rather than a settled technological breakthrough.
Will quantum computers replace normal computers?
Probably not.
Quantum computers are being developed for particular classes of problems where quantum algorithms may provide advantages. Classical computers will remain extremely useful for ordinary tasks such as web browsing, office software, gaming, databases, and most everyday computing.
The future is more likely to involve quantum-classical hybrid computing than a complete replacement of classical machines.
Does quantum computing threaten online security?
Potentially, yes.
A sufficiently capable quantum computer could threaten some widely used public-key cryptographic systems. That’s why NIST finalized three post-quantum cryptography standards in 2024 and is encouraging organizations to begin migrating to quantum-resistant algorithms.
Conclusion: Where Quantum Computing Goes From Here
The latest quantum computing breakthroughs show that the field is moving beyond impressive laboratory demonstrations and toward the much harder problem of building reliable, scalable machines.
Google’s Willow processor demonstrated an important step in quantum error correction. IBM’s Heron systems pushed superconducting hardware toward deeper and faster circuits. AI is being used to improve error decoding, while researchers are exploring photonic, topological, and other approaches to building quantum processors.
And perhaps most importantly, researchers are beginning to place greater emphasis on logical qubits, verification, and useful quantum advantage rather than simply asking how many physical qubits a machine contains.
The latest breakthroughs in quantum computing 2024 were important because they helped demonstrate that error correction and scaling could work in ways researchers had been trying to achieve for decades. Developments in 2025 and 2026 have continued that trajectory, while also showing how many technical and scientific questions remain open.
Quantum computing isn’t ready to replace your laptop—and anyone claiming otherwise is getting ahead of the evidence.
But the technology is no longer purely theoretical either.
The next major milestone won’t simply be another bigger quantum processor. It will be a machine with enough reliable logical qubits, strong error correction, and useful algorithms to solve a real-world problem that conventional computers cannot efficiently handle.
That’s the breakthrough worth watching.
