11-06-2025
Why is big tech fussing over fault-tolerant quantum computers?
For more than a decade, big tech companies like IBM, Google and Microsoft have been loudly announcing that quantum computers will soon solve tasks that stump today's supercomputers. From simulating molecules for drug discovery to optimizing global logistics, they promised revolutionary breakthroughs, and all that in seconds instead of months and years.
Quantum computers do hold a lot of promise. Two bits in the traditional or classical computers we use in homes and offices today can represent one of four possible states—00, 01, 10, or 11—but only one at a time. In contrast, two quantum bits (qubits) in a quantum computer can represent all four states simultaneously, thanks to superposition and entanglement properties, effectively functioning like four classical computers in one.
However, as a quantum computer's power grows exponentially when you add more qubits, it also becomes more prone to errors. In technical terms, these are known as Noisy Intermediate-Scale Quantum (NISQ) devices that typically host fewer than a few hundred qubits, but are incredibly fragile and lose their quantum properties when disturbed by variables like heat, vibrations, or electromagnetic interference.
Tech companies use the term "error correction" to remedy this situation. It can be likened to having multiple backup musicians playing the same part in a noisy concert hall. If one musician makes a mistake, the others can detect it and correct it in real-time. In quantum computing, this means using many physical qubits (individual musicians) to create one "logical qubit" (the perfect musical note) that can maintain its quantum state reliably.
But how soon is soon?
Quantum, by its very nature, is unpredictable. Hence, the error correction process is gradual and not a one-time achievement. For instance, Google's quantum computing chip named Willow made waves in December 2024 for its ability to reduce system errors despite adding qubits, and solving a computation in under five minutes that would take a supercomputer 10 septillion years (1 with 25 zeroes), more than the age of the Universe, to finish. It even prompted Elon Musk to react with a 'Wow' when Google CEO, Sundar Pichai, announced this on X.
But the next challenge, as Google itself put it, was "to demonstrate a first 'useful, beyond-classical' computation on today's quantum chips that is relevant to a real-world application". Google is targeting 1 million qubits by the end of the decade, though error correction means only 10,000 will be available for computations.
This February, Amazon Web Services (AWS) announced its new Ocelot quantum computing chip "that can reduce the costs of implementing quantum error correction by up to 90%". AWS designed Ocelot's architecture with built-in error correction from the ground up, using 'cat qubits'—named after Schrödinger's cat—for their natural ability to suppress certain errors.
In a first, AWS researchers integrated cat qubits with additional error correction components on a microchip, using scalable manufacturing techniques adapted from the microelectronics industry.
That very month, Microsoft too announced a "significant leap in quantum computing" with the launch of Majorana 1, which it touted as a revolutionary quantum chip powered by a new topological core. While introducing the chip, Satya Nadella said on X that this "entirely new state of matter (the other three main states of matter being: solid, liquid and gas)", unlocked by a new class of materials called "topoconductors", is what powers Majorana 1. Microsoft added that this means the chip incorporates error resistance at the hardware level, making it inherently more stable.
IBM advances the timeline
IBM had earlier announced its commitment to building a 100,000-qubit fault-tolerant system by 2033. On 10 June, it advanced this deadline, unveiling a roadmap to build the world's first large-scale, fault-tolerant quantum computer by 2029.
"IBM is charting the next frontier in quantum computing," CEO Arvind Krishna said in a press statement, attributing the progress to the company's "expertise across mathematics, physics, and engineering is paving the way for a large-scale, fault-tolerant quantum computer that will solve real-world challenges".
The IBM Quantum Starling, to be housed in a new data center in New York, will perform 20,000 times more operations than current quantum computers. Its computational state would require memory equivalent to more than a quindecillion (1 and 48 zeroes) of today's most powerful supercomputers.
Further, the new system addresses quantum computing's fundamental challenge: error correction. Starling will execute 100 million quantum operations using 200 logical qubits—units that combine multiple physical qubits to monitor and correct errors.
This foundation will enable IBM Quantum Blue Jay, capable of 1 billion operations over 2,000 logical qubits. IBM's breakthrough centres on quantum low-density parity check (qLDPC) codes, which reduce required physical qubits by approximately 90% compared to existing methods. It has detailed the architecture's efficiency and real-time error correction capabilities in two separate technical papers.
The roadmap includes three stepping stones: IBM Quantum Loon (2025) will test qLDPC architecture components; Kookaburra (2026) combines quantum memory with logic operations as the first modular processor; and Cockatoo (2027) will link multiple quantum chips together.
This fault-tolerant quantum computer could revolutionize drug development, materials discovery, chemistry, and optimization by accessing computational power previously impossible to achieve at scale. These advancements are designed to culminate in Starling in 2029.
'IBM's decision to pull its 100-000-qubit goal forward to 2029 is just one out of multiple announcements to be made for quantum computing's commercial decade," Anders Indset, business philosopher and tech investor, said.
'Capital is pouring in, talent is converging, and the first consolidations are already reshaping the field. When money and competence collide at this scale, progress doesn't follow Moore's law (number of transistors on a computer chip doubles roughly every two years, making computers faster and cheaper over time), it can be much more radical."
Indset expects more breakthrough announcements this year. "Boards that still treat quantum as a slide-deck talking point and potential future topics will soon be negotiating with rivals who are quantum secure and can model molecules or optimise global supply chains in minutes, not months," he added.
What are other companies doing?
Precedence Research pegs the global quantum computing market at $1.44 billion in 2025 and predicts it to touch $16.22 billion by 2034. While IBM, Google, Microsoft, and AWS lead with hardware-heavy approaches, startups like Quantinuum, PsiQuantum, and Riverlane are pushing complementary innovations—from new codes to scalable photonics and decoding software.
D-Wave doesn't currently use standard quantum error correction because its machines are annealers, not universal quantum computers. Instead, it focuses on noise-aware design and error mitigation. A quantum annealer is a special type of quantum computer designed to solve optimization problems (finding the best solution out of many possible ones).
Closer home, India's National Quantum Mission (NQM), launched in 2023 with an outlay of ₹6,003 crore, has already awarded up to ₹30 crore each to eight startups—QNu Labs, QPiAI India, Dimira Technologies, QuPrayog, Quanastra, Quan2D, Pristine Diamonds, and Prenishq. The idea is to jumpstart India's presence in quantum technologies, including quantum computing, sensing, communications, and quantum error correction (QEC).
With NQM, India plans to develop quantum computers with 50-100 qubits in about five years, and accelerate it to 1000 qubits and beyond in eight years. Recognizing the importance of robust algorithm development—particularly for fault tolerance and error correction—Ajai Chowdhry, chairman of the NQM Mission Governing Board, and EPIC Foundation, told Mint in an interview that NQM is launching a dedicated initiative focused on quantum algorithms.
The adoption of quantum technologies across industries could potentially add $280–310 billion of value to the Indian economy by 2030 with the manufacturing, high-tech, banking, and defence sectors at the forefront of quantum-led innovation, according to a 2022 Nasscom-Avasant report.
