The Quantum Computing Revenue Story Is Real. The Hardware Comparison Is Not.
IonQ reported $64.7 million in first-quarter revenue this month — 755 percent year-over-year growth and the largest quarterly revenue in the company's history, according to a GNews Quantum wire. The company also opened a 22,000-square-foot quantum R&D laboratory in Boulder, Colorado, in May, expanding its hardware development footprint. The financials look real. The question is what sits underneath them.
IonQ holds a world record. In October 2025, the company achieved 99.99 percent two-qubit gate fidelity — the highest ever measured for a quantum computer — surpassing the previous 99.97 percent record held by Oxford Ionics, which IonQ acquired earlier that year. The performance jump from 99.9 percent fidelity to 99.99 percent translates to a 10 billion times improvement in error rate on the same hardware, meaning calculations that would fail routinely at the lower threshold run reliably at the higher one. IonQ says it is on track to deliver 256-qubit systems in 2026, with prototypes already in its labs, and a roadmap to millions of qubits by 2030. These are verifiable claims backed by a public benchmark.
Microsoft's case is different. Its Majorana 1 chip, announced in February 2025, uses a class of materials the company calls a "topoconductor" to implement topological qubits — an architecture that, if it works at scale, would be more error-resistant than trapped-ion approaches. Microsoft has published a roadmap targeting a million qubits on a single chip. The hardware currently exists and has eight qubits. And those eight qubits have so far only demonstrated a readout method — the chip can indicate whether it is in one quantum state or another, but there has been no published coherence test showing the device can perform a quantum computation and maintain that state long enough for the result to be useful. The history here is relevant: Microsoft had a Majorana-related paper retracted from Nature in 2018 after other researchers identified errors in the data.
The market is treating these as equivalent stories — two quantum companies, both underappreciated, both on the verge. They are not equivalent today. IonQ has demonstrated a hardware milestone with an independent prior benchmark and is booking revenue against it. Microsoft has announced a materials approach with a published roadmap and an existing chip that has only shown it can read out a quantum state, not process one. These are two different kinds of claim: one is evidence, the other is an aspiration with a business card.
What would change this picture for Microsoft is a demonstration of actual quantum processing on the Majorana 1 chip — not just readout, but a coherence test showing the device can hold and manipulate a quantum state through a computation. If that happens, the comparison becomes more legitimate. If it doesn't, the distinction between "we have built something that demonstrates the physics we think will work" and "we have built something that computes" remains. IonQ is in the second category today. Microsoft is in the first.
The broader point isn't to dismiss topological qubits as an approach. If they work at scale, they would represent a genuinely different and potentially more scalable architecture than trapped ions. The point is that these two companies are not comparable investments today based on what each has actually demonstrated. The market treating them as equivalent may say more about how quantum press releases are consumed than about the actual state of the hardware.