Why IBM predicts quantum advantage within two years

Industry analysts from McKinsey to Omdia largely converge on a timeline for initial quantum advantage emerging in the next few years. While the era when quantum computers can routinely tackle large-scale challenges in fields like drug discovery and materials science might still be years away, IBM’s Quantum CTO, Oliver Dial, Ph.D., predicts the threshold of ‘quantum advantage’—where quantum machines outperform classical ones on specific, useful tasks—is closer than many think. “We think using techniques like error mitigation, in the next two years we will actually be showing quantum advantage,” Dial stated in a recent interview. He foresees a future that taps current hardware capabilities even before fully fault-tolerant machines arrive.

IBM’s Gross code ‘game changer’

Dial’s confidence echoes a period of accelerating progress across the quantum landscape, with competitors like Google also reporting significant advances in error correction and collaborations like Quantinuum with JPMorgan Chase recently demonstrating potential real-world applications with random numbers. Against this backdrop, IBM highlights an advance: a novel “Gross code” approach to quantum error correction. The term “gross” refers to the quantity 144 (a dozen dozen), reflecting a key aspect of the code’s structure. According to Dial, this method represents a “total game changer.” “The Gross code bought us two really big things,” Dial explained. “One is a 10-fold reduction in the number of physical qubits needed per logical qubit compared to typical surface code estimates.” This reduction is significant for practical construction. “From an engineering perspective, that’s huge,” Dial elaborated, noting that dealing with fewer inherently unreliable components, from connectors to the qubits themselves, means “the engineering of building this system got enormously easier.”

Dial highlighted the divergence in perspectives on achieving near-term results:

We think we’re going to have quantum advantage in two years using these techniques. The contrast between that and what you were talking about before is striking.

Near-term advantage and long-term fault tolerance

This dual focus—pushing near-term gains while building for the long haul—is important. The “quantum advantage” Dial anticipates within two years hinges on error mitigation. He underscored the significance of this near-term strategy, contrasting it with approaches solely focused on future fault tolerance: “We think we’re going to have quantum advantage in two years using these techniques,” he said.

These mitigation techniques, he described, are ways to “recover accurate answers from computers that commit errors.” That, in turn, allows useful work on today’s inherently “noisy” quantum processors. Yet tackling truly large-scale simulations requires fault tolerance—systems that actively correct errors. This long-term goal is accelerated by the specific type of quantum Low-Density Parity Check (qLDPC) code IBM is pursuing, which Nature detailed in 2024. As Dial explained, this qLDPC approach allows a block of 12 stable logical qubits to be encoded and protected using only about 300 physical qubits, drawing on specific, potentially complex qubit connectivity patterns.

This design represents a considerable improvement over earlier methods often requiring thousands of physical qubits for a single logical qubit. This efficiency gain is important for achieving the goals laid out in IBM’s public quantum roadmap, which is a multi-year plan detailing stepping stones like enhancing quantum execution speed and parallelization in 2025, progressively increasing circuit depths to 10,000 gates by 2027, and culminating in the delivery of a fully error-corrected system with 200 logical qubits capable of running 100 million gates by 2029.

The Gross code advance significantly boosts the feasibility of hitting this demanding 2029 target. While IBM remains committed to achieving these specific roadmap milestones, Dial stressed the team retains flexibility on the exact technical path used to get there: “This is a roadmap, not a death pact,” he stated. “If circumstances change, we’ll still accomplish these goals, but we may do it differently.”

This shift towards more efficient codes like the Gross code addresses significant engineering hurdles associated with earlier concepts. Scaling surface codes, Dial noted, often implied massive, potentially “football field-sized arrays” of thousands or tens of thousands of physical qubits just to create a small block of reliable logical qubits. Managing the wiring and yield for such monolithic structures presented significant practical challenges. In contrast, the Gross code’s ability to package 12 logical qubits within a roughly 300-physical-qubit module creates a “repeating unit cell” that is feasible to build, test, and integrate with current and near-term fabrication capabilities, significantly de-risking the path to the 2029 goal.

The team behind the quantum push: ‘Our life’s work’

Executing the quantum roadmap naturally requires long-term commitment. Dial points to the team’s longevity as a key factor: “I’ve been on this team for 12 years. Almost all the original team from 12 years ago is still here because this is our life’s work—to get these machines out the door.” This shared sense of purpose provides clarity, especially within a large organization. IBM. “IBM is a big place,” Dial noted. “It’s hard to describe what our products are or exactly what IBM does in less than 50 sheets of paper. But ‘build the world’s first useful quantum computers.’ That’s something I don’t really need to sell anybody on.”

That public declaration of intent sets a high bar for execution. “Now that we’ve put it on the roadmap, that’s when the real work begins,” Dial acknowledged. “We still have to make that happen.”

“One thing I love about working on quantum computing is it’s the easiest thing in the world to get people excited about,” Dial reflected. “I could go talk to your grandmother and get her excited about quantum computing just by saying those words.”