Quantum Computing Nears Critical 100 Qubit Milestone
From Clunky
Beginnings to a Measurable Goal
The Electronic Numerical Integrator and Computer, known as Eniac, was a monster of its time. Built in 1946 by the University of Pennsylvania for the U.S. Army, it weighed roughly 30 tons. Its 18,000 vacuum tubes filled an entire room, and it operated at about 5,000 hertz, performing 5,000 additions per second. Adjusted for inflation, the cost to build Eniac was around $9 million. Today, a typical laptop handles more than 3 billion calculations per second and costs roughly $1,000, a fraction of Eniac’s inflation-adjusted $9 million price tag.
Technology follows a familiar pattern: it starts out clunky, slow, and expensive, then improvements accelerate as it becomes more useful. Quantum computing sits in a pre-Eniac stage now, but an important milestone is approaching, and dozens of large companies are preparing for it. Nvidia Corp.‘s classical computer chips can simulate quantum performance without the technical drawbacks of actual quantum machines
Quantum computers promise to accelerate drug discovery, design new materials, optimize logistics routes, improve financial risk management, and reshape encryption and cybersecurity. Before any of that becomes practical, they must overcome the error problem. Classical computers use digital bits that process information in 1s and 0s via electric current switching tiny transistors on and off. Quantum computers use quantum bits, or qubits, which can hold the values of both 0 and 1 simultaneously — a state known as superposition.
The power of qubits scales exponentially with additional qubits, unlike classical bits, which run operations sequentially. Qubits explore different potential states simultaneously, enabling quantum machines to handle far more complex calculations. However, qubits are extremely delicate — vibrations, heat, and even cosmic rays can collapse the quantum state and produce errors.
Two Roads to the Same Destination
To counter qubit fragility, quantum computers need error-corrected qubits, also known as logical qubits. Information is encoded across several physical qubits to provide redundancy, allowing the system to identify and correct errors. Future quantum computers may have tens of thousands of logical qubits, but scaling these delicate machines while controlling errors will take time and possibly unforeseen breakthroughs. Most experts agree that 100 error-corrected qubits is the threshold for practical usefulness.
Quantinuum plans to release its Sol computer in 2027 with 100 logical qubits, and IBM announced it achieved 70 logical qubits last month. Quantinuum plans to release its Sol computer in 2027 with 100 logical qubits. [3] This machine tops its Helios model, which is available now and has 48 error-corrected qubits. International Business Machines Corp. announced last month that it achieved 70 logical qubits. [4] That experiment was conducted with the University of Chicago. [5] Infleqtion expects its latest design to provide more than 50 logical qubits next year as it scales toward 100. Reaching 100 logical qubits is the point where quantum computers start to solve important problems that classical computers cannot.
Pranav Gokhale, chief technology officer and co-founder of Infleqtion, said during his company’s earnings call that reaching 100 logical qubits is the point at which quantum computers start solving important problems in material science and chemistry, with potential extensions into artificial intelligence. Distinct approaches to building quantum computers make tracking industry progress tricky. Some designs use mechanical qubits etched and wired together on a chip. These human-made qubits scale more easily than other designs but have higher error rates from initial imperfections and heat from connecting wires.
Other designs use natural particles such as atoms or photons. Atoms are trapped with magnetic fields and manipulated with lasers; photonic processors use beam splitters and phase shifters to control photons. These qubits start out perfect, but errors appear when they are moved, making these machines harder to scale than chip-based designs. Infleqtion uses neutral atoms, Quantinuum works with charged atoms known as ions, and PsiQuantum’s qubits are based on photons
IBM, Google, and Rigetti Computing Inc. all use mechanical qubits in their designs, etched and wired together on a chip. [7] These designs benefit from easier scaling compared to natural particle approaches. The initial imperfections and heat from connecting wires create higher error rates, though. The natural particle approaches start with perfect qubits but introduce errors during movement. Each pathway has its own set of trade-offs between scalability and error rates. It is too early to tell which of these pathways will be the winner in the end. Multiple solutions may ultimately prove viable, each serving different computational needs.
A Shift in the Conversation Signals Arrival
In May, the U.S. Commerce Department selected nine companies for $2 billion in funding under the CHIPS and Science Act, including IBM, Rigetti, Quantinuum, Infleqtion, and PsiQuantum. The selected companies include IBM, Rigetti, Quantinuum, Infleqtion, and PsiQuantum. This diversity of funding acknowledges that multiple approaches may prove viable. The government support also signals that quantum computing has moved beyond theoretical curiosity. The investment represents a significant commitment to an emerging technology with uncertain outcomes. Each company brings a different technical approach to the same fundamental challenge of building a usable quantum computer.
The conversation around quantum computing has shifted from whether the technology will ever be valuable to how to build error-corrected logical qubits at scale. The remaining challenges are engineering problems rather than fundamental physics questions. When a machine offers 100 logical qubits or more, the quantum-computing age will have arrived.
The path from Eniac to the modern laptop took decades of incremental improvement and occasional breakthroughs. Quantum computing appears to be on a similar trajectory, though possibly accelerated. Companies are racing toward the 100-logical-qubit threshold with different technical approaches, and which design will dominate remains an open question.
Scaling beyond 100 logical qubits toward the tens of thousands needed for full-scale quantum computing will require solving the error problem with approaches not yet fully developed. Each technical pathway faces distinct scaling challenges, and the timeline for reaching practical, large-scale quantum machines remains uncertain.
Sources
2. Nvidia Corp.
3. Quantinuum
4. International Business Machines Corp.
6. PsiQuantum
7. Google