Quantum's New Trade: QUBT Spends Its Treasury on Chip Capacity, Not Qubits

by Arash Shirazi · August 13, 2026

Quantum Computing Correspondent, Frontier Tech Wire

Quantum Computing Inc. (Nasdaq: QUBT) published second-quarter results on August 10 that read, at first glance, like an ordinary development-stage quarter: $5.6 million of revenue against $21.8 million of operating expenses. The line that matters more sits on the balance sheet. The company closed June with roughly $1.3 billion in cash, cash equivalents and investments, and it is deploying a slice of that hoard not on qubits but on semiconductor manufacturing capacity. Ten days earlier, on July 31, IonQ (NYSE: IONQ) made the same trade on a far larger scale. Vertical integration into chips, rather than any hardware milestone, has become the defining capital-allocation theme of the quantum sector's reporting season.

The top-line growth is real but starts from almost nothing. QUBT's $5.6 million compares with $61 thousand in the second quarter of 2025 and $3.7 million in the first quarter of this year. Operating expenses rose 114 percent year over year to $21.847 million: $8.428 million of research and development, $1.932 million of sales and marketing and $11.487 million of general and administrative costs. The G&A line carries $7.3 million of acquisition-related transaction expenses, so underlying overhead is lower than the headline implies — but operating costs still run at roughly four times revenue.

The reported net loss of $11.8 million, or five cents per basic share, compares with a $36.5 million loss a year earlier — but that improvement is an accounting artifact rather than an operating one. The prior-year quarter absorbed a $28.0 million mark-to-market loss on warrant derivative liabilities; this quarter's equivalent charge was $1.7 million. Strip out both non-cash swings and the picture inverts: operating expenses grew faster than revenue and cash burn widened. A narrowing GAAP loss here is not evidence of operating leverage.

A second oddity deserves naming plainly. Interest and other income totaled $13 million in the quarter — up from $1.8 million a year earlier — against $5.6 million of revenue, meaning QUBT's treasury currently generates more than twice as much income as its products do. That swing accounts for much of the rest of the year-over-year change in the bottom line. Total liabilities were just $47.2 million against $1.59 billion of stockholders' equity, so the position is comfortable — but the reported bottom line is unusually sensitive to interest rates and to how fast management spends the pile down.

Spending it down is exactly what management is doing. QUBT completed the acquisition of NHanced Semiconductors for $73.1 million in initial consideration, structured as cash and stock, with a conditional earn-out of up to $72 million; combined with earlier deal activity, roughly $180 million of cash went to acquisitions in the first half. The NHanced assets became what QUBT calls Fab 2, launched ahead of schedule, adding advanced packaging to the thin-film lithium niobate photonics work the company already does. On the earnings call, executives described the facility as running at around 60,000 wafers annually, flagged upgrade capital expenditure of $50 million to $100 million with roughly $75 million as the working estimate — while cautioning that the plan is still at the planning stage and that little of that money is expected to be spent in 2026 — and put the acquired business's expected annual revenue contribution at $7 million to $16 million depending on project delivery.

The rest of the commercial update was incremental. Contract backlog stood at $42.5 million, which management said should underpin operations for 12 to 18 months. A Dirac-3 optimization system was sold and installed at a global consulting firm, and a framework agreement with Planck Dynamics covering the NeuraWave photonic platform carries potential value above $10 million — potential being the operative word, since framework agreements are not firm orders. Executives put 70 to 80 percent of the business in government contracting, which makes revenue lumpy and delivery timing hard to forecast.

The market's response was strikingly muted for a name that has historically traded on sentiment. Simply Wall St noted the shares moved just 0.06 percent on the results, sitting near $8.94, and a separate analysis published August 11 put the one-year return at roughly negative 46 percent. For a company that averaged about 224.7 million basic shares in the quarter and built its cash position through dilution, the flat reaction reads as investors treating the revenue ramp as necessary but not yet sufficient.

The IonQ comparison sharpens the theme. IonQ closed its acquisition of SkyWater Technology on July 31, paying $15.00 in cash plus 0.4883 IonQ shares for each SkyWater share in a deal valued at roughly $1.8 billion; SkyWater now operates as a wholly owned subsidiary under chief executive Thomas Sonderman, reporting to IonQ chief executive Niccolo de Masi. IonQ's own second quarter, reported August 5, showed revenue of $80.1 million, up 287 percent, with full-year guidance raised to $280 million to $290 million from $260 million to $270 million.

IonQ's headline GAAP net loss of $1,867.7 million, or $5.08 per share, is the sector's clearest illustration of why that line needs unpacking. Roughly $1.6 billion of it came from a non-cash change in the fair value of warrant liabilities — a remeasurement that moves with the share price and consumes no cash. Adjusted EBITDA loss was $120.3 million, or $95.6 million excluding SkyWater-related spending incurred before closing. IonQ ended June with $3.0 billion in cash, equivalents and investments, around $2.0 billion pro forma after funding the deal, and no debt. Shares closed down 4.29 percent at $39.93 on the day of the report.

The strategic logic behind both deals is the same: photonic and trapped-ion architectures depend on specialized chip fabrication and packaging that no merchant foundry prioritizes, and owning the line shortens iteration cycles while removing a supplier chokepoint. The counterargument is equally straightforward. Fabs are capital-hungry, cyclical and management-intensive, and two companies with combined product revenue under $90 million a quarter have just absorbed the operating complexity of semiconductor manufacturing without yet demonstrating that owning silicon accelerates the quantum roadmap.

A housekeeping note for anyone working from older material: listing facts here have shifted repeatedly. D-Wave Quantum announced on July 14 that it was moving from the New York Stock Exchange to Nasdaq while keeping the ticker QBTS, a transfer that took effect after the close on July 24 with Nasdaq trading beginning July 27, and Quantinuum began trading on the Nasdaq Global Market under QNT on June 4 after pricing an upsized offering at $60 per share for roughly $1.68 billion. IonQ remains a New York Stock Exchange listing.

The risks are conventional for the category and should not be glossed over. Operating costs run at multiples of revenue; government concentration makes results unpredictable; the NHanced earn-out and the roughly $75 million of planned facility upgrades are further claims on a treasury that fell from about $1.5 billion at the end of 2025 to $1.3 billion by June 30; and the share base reflects substantial past dilution. What to watch is testable: whether Fab 2 begins contributing revenue at a pace consistent with the $7 million to $16 million annual range management indicated, whether the backlog converts on schedule, and whether operating expenses stabilize once acquisition costs roll off.