The Announcement That Broke the Internet
Last week, the Department of Commerce dropped a bombshell: $2.013 billion in federal incentives for nine quantum computing companies, structured as equity investments rather than traditional grants. The money comes from the CHIPS and Science Act — the 2022 bill designed to boost domestic semiconductor manufacturing. Except quantum computing isn't really semiconductor manufacturing, at least not in the way Congress imagined it.
The breakdown is staggering. IBM gets $1 billion to spin up Anderon, a new quantum foundry subsidiary. GlobalFoundries takes home $375 million for its own quantum fabrication ambitions. The rest splits among seven startups: D-Wave, Rigetti, Infleqtion, PsiQuantum, Quantinuum, Atom Computing, and Diraq — each receiving between $38 million and $100 million.
Here's what makes this genuinely unusual: the government isn't writing checks and walking away. It's taking minority, non-controlling equity stakes in every company. The same pattern we saw with Intel (10% stake, making the federal government its third-largest shareholder) and MP Materials. The Trump administration has been building this equity-for-support model since day one, but quantum is the biggest bet yet.
Markets loved it. IBM jumped 12%. D-Wave surged 33%. Rigetti soared 30%. Infleqtion skyrocketed roughly 31%. Even companies not in the deal — IonQ, Arqit, Quantum Computing Inc. — rode the wave higher.
But not everyone was celebrating.
The Legality Question Nobody Can Ignore
Zoe Lofgren, the ranking Democrat on the House Science, Space, and Technology Committee, didn't mince words. "This announcement is illegal and troubling on so many levels," she said the day after Commerce revealed the deals.
Her objection rests on three pillars, and any of them alone might be enough to sink the program in court.
First, the CHIPS and Science Act was explicitly designed for microelectronics R&D with a focus on classical semiconductor technology. Quantum processors use some overlapping fabrication techniques — lithography, thin-film deposition, cleanroom processes — but the underlying physics is entirely different. Superconducting qubits don't behave like transistors. Trapped ions don't care about Moore's Law.
Second, the law was structured to foster public-private research partnerships. These deals aren't partnerships. They're investments. The government gets equity. Companies get cash. There's no university involved, no National Lab running parallel experiments, no open research agenda.
Third — and this is the politically charged one — Lofgren pointed out that Dario Gil, a former IBM executive who now serves as Under Secretary for Science at the Department of Energy, was involved in negotiating these deals. That's not necessarily corruption. But it's the kind of revolving-door detail that makes congressional oversight committees twitchy.
A lawsuit would require standing — some entity harmed by the fund diversion. A university that lost out on a CHIPS partnership grant might have that standing. But by the time any court case reached a verdict, the money would already be spent. The deals are structured as letters of intent, not binding commitments, but the political momentum is real and the companies have already started announcing plans.
The legal ambiguity extends beyond statutory interpretation. If quantum computing advances faster than expected, the national security implications are profound — particularly for cryptography. Nations that deploy quantum computers first could break current encryption standards, a threat already being addressed in places like Japan, which has deployed the world's first large-scale quantum key distribution network. See: Japan Launches World's First Large-Scale Quantum Key Distribution Network.
Anderon: America's First Quantum Foundry
The crown jewel of the announcement — and its most controversial element — is Anderon. Named after a play on "and" (the company's website spells it without the leading capital), this new IBM subsidiary will operate as a pure-play quantum wafer foundry headquartered in Albany, New York.
The model is straightforward: IBM contributes $1 billion in cash, the government matches with $1 billion from CHIPS funds, and IBM transfers significant intellectual property, fabrication assets, and a skilled workforce to the new entity. Anderon will fabricate 300-millimeter quantum wafers — specifically for superconducting qubit architectures — and sell foundry services to other quantum companies.
Think TSMC, but for quantum chips instead of classical processors. That's the analogy IBM is pushing, and it's not entirely wrong. A significant number of companies are designing transmon-based quantum hardware that differs from IBM's approach in important ways. But they're stuck producing limited test samples in fabs that don't specialize in quantum hardware, or competing with academic users for fabrication time. Anderon would give them access to proper quantum-grade manufacturing.
There's a problem with this analogy, though. TSMC serves multiple competing chip architectures because classical computing has converged on a few standard platforms. Quantum computing hasn't converged at all. Transmons dominate right now, but so do trapped ions (Quantinuum), neutral atoms (Atom Computing, Infleqtion), photonic systems (PsiQuantum), silicon spin qubits (Diraq), and even topological approaches that don't use hardware in the conventional sense. Anderon will initially serve only superconducting qubits.
That means government money is being used to bet on one specific quantum technology over all others. If trapped ions or photonic computing turns out to be the path to practical quantum advantage, Anderon's entire value proposition shrinks. You don't get that kind of asymmetry with a $2 billion bet.
Anderon's location in Albany is no coincidence — it sits in IBM's existing semiconductor research ecosystem, the same region where IBM has been developing its next-generation classical chip technology. For context on IBM's broader fabrication ambitions beyond quantum, see: Reaching for the Z-Axis: Inside IBM's 7-Angstrom 3D Chip Breakthrough.
The Portfolio Play Across Quantum Modalities
Where the Commerce Department's approach gets interesting is in how it spread money across competing quantum technologies. The seven startup awards aren't concentrated in one modality:
Atom Computing ($100M) is building neutral-atom systems that manipulate thousands of qubits using optical tweezers. Infleqtion ($100M) is also in neutral atoms but focused on the optical systems and error correction needed to scale them. PsiQuantum ($100M) is going all-in on photonic quantum computing — using light instead of superconducting circuits. Quantinuum ($100M) is scaling trapped-ion systems with advanced photonics. D-Wave ($100M) is advancing both annealing and gate-model superconducting approaches. Rigetti ($100M) is developing next-generation superconducting architectures with integrated cryogenic electronics. Diraq ($38M) is working on silicon spin qubits, which could eventually leverage existing semiconductor fabrication infrastructure.
This portfolio approach makes strategic sense. We're years away from useful error-corrected quantum computing, and closer to a decade from tackling the problems where quantum computers could see widespread commercial use. Nobody knows which modality will win. Keeping all of these companies viable ensures the U.S. doesn't miss the breakthrough whenever it comes.
But viability and profitability are different things. Several of these companies are likely years away from revenue that justifies their valuations, let alone returns for the government's equity stakes. The boom-bust cycle in quantum hardware is already visible: Anderon's target market for superconducting wafers will be concentrated in a handful of data centers once error-corrected systems go operational, since transmons require milliKelvin temperatures and massive refrigeration infrastructure. Annual demand for quantum wafers could be far smaller than the $850 billion economic value projection that IBM and Commerce are citing for 2040.
The Bigger Pattern: Government as Venture Capitalist
This quantum announcement didn't happen in a vacuum. It's the latest chapter in a pattern that started when the Trump administration took office in 2025.
The Pentagon made MP Materials — a rare earths mining company critical to defense supply chains — its largest shareholder, explicitly to counter China's dominance. Intel received a 10% government stake, making the federal government its third-largest shareholder. Now quantum computing gets the same treatment.
The logic is seductive: when a technology matters for national security, why let private markets alone determine its fate? Why not take equity, ensure domestic control, and capture returns for taxpayers when these companies succeed?
The problem is that equity stakes change incentives. Companies receiving government money with equity conditions face pressure to prioritize government objectives over shareholder value. They also face the political risk of being perceived as too close to the administration that invested in them — a problem for companies that might want to sell quantum computing services to foreign governments or cooperate with international research consortia.
And then there's the question of exit. When the government eventually sells its stakes — and it will, sooner or later — who buys them? If quantum computing companies succeed beyond expectations, the government captures significant returns. If they fail, taxpayers absorb the loss. It's heads-private, tails-public in a way that traditional grants avoid.
The CHIPS and Science Act was supposed to be about rebuilding American semiconductor manufacturing. What we got instead is a new model of industrial policy where the government doesn't just fund research — it becomes a venture capitalist, picking winners, taking equity, and betting billions on technologies that may or may not work.