Quantum Snake Oil: A Field Guide to Misleading Quantum and PQC Technology Marketing

This is the capstone article of the Quantum Snake Oil Dictionary, a Deep Dive series examining terms used in quantum technology marketing. I published it in May 2026 and updated it in October 2026 to cover all 35 entries.

Quantum Snake Oil From 2001 to 2026

I first wrote about quantum snake oil in 2001, when a startup was marketing “neural quantum encryption” that turned out to be a rebranded XOR cipher. Back then, a handful of vendors were slapping “quantum” onto products like that and calling it innovation. The audience was small, the stakes were modest, and most security professionals could smell the nonsense from across the trade show floor.

Twenty-five years later the problem has metastasized. Companies add “quantum” to their pitch to raise venture capital and win government contracts. Fraudsters add it to investment schemes, and financial regulators in more than a dozen countries have flagged “Quantum AI Trading” platforms that involve no quantum computer at all. Vendors tell CISOs that RSA is already broken, although the largest number anyone has factored with Shor’s algorithm on real quantum hardware is 21. Others sell chaos-based ciphers as quantum-proof 37 years after the first published critique of one, or announce “zero logical errors” without saying how many runs they discarded to reach zero. In June 2026 I counted three extraordinary quantum claims in two weeks, none of them peer-reviewed and all of them tied to commercial events.

The gap between what quantum technology can do and what marketing departments claim for it is the widest I have seen in 25 years of watching this market. A legitimate quantum security product can be described with an algorithm name, a standard, a certificate number and a stated threat model. Most of the terms in this dictionary are used in place of that description.

Quantum computing is real. Post-quantum cryptography (PQC) is real, and NIST published its first three post-quantum standards in August 2024. The threat to current encryption is real. Vendors borrow credibility from those real results, and buyers who cannot check the physics themselves have no way to see where the results end and the marketing begins. Many companies that use quantum terminology are perfectly honest, which makes the dishonest ones harder to spot.

I have covered this problem from several angles on PostQuantum.com. My Quantum of Flapdoodle maps the full spectrum of quantum misinformation, from harmless hype inflation through willful marketing spin to outright financial fraud and pseudoscience. Q-FUD: The Quantum Panic Industry examines how vendors turn fear, uncertainty and doubt (FUD) about the quantum threat to encryption into sales. The Quantum Baloney Detection Toolkit offers practical heuristics for evaluating claims. On the denial side, I have explored the spectrum from healthy skepticism through performative cynicism to bad-faith contrarianism exploited for profit, and in May 2026 I examined how denial and grift form a symbiosis that paralyzes real engineering, using Bitcoin’s response to the quantum threat as a case study. The Quantum Winter Warning shows what happens when the hype boomerangs and damages the entire field.

The Quantum Snake Oil Dictionary works term by term. Each entry takes one piece of quantum marketing vocabulary and tests it against the physics, the mathematics and the standards. It closes with a verdict and the questions I would put to the vendor. When a CISO, an investor or a procurement officer hears one of these terms in a sales pitch, they should be able to find it here and know what to ask next.

Why a Dictionary?

The first reason is search. When a vendor says its product uses “simulated quantum entanglement” or is “quantum-safe certified,” the first thing many buyers do is search for the phrase. If a physics-based analysis ranks for it, the buyer finds the analysis before signing the contract. A broad overview article, however well written, will not rank for 35 specific phrases. A dictionary entry for each phrase will.

The second reason is education by accumulation. Most entries take about five minutes to read, and most follow the same shape: a note on scope, what the term claims, what the physics, mathematics or standards say, the questions to ask a vendor and a bottom line. After ten entries a reader starts to see the same rhetorical moves across different terms, vendors and market segments. Recognizing the move is more useful than debunking any single term, because the next term a vendor coins will reuse one of the old moves.

Bruce Schneier listed nine warning signs of cryptographic snake oil in his Crypto-Gram newsletter in February 1999. Several of them reappear in this dictionary with quantum vocabulary attached, among them ridiculous key lengths, one-time pads, proprietary cryptography and cracking contests.

Where to Start, by Role

Thirty-five entries is a lot to read before a vendor meeting. These are the ones I would read first.

• Buying encryption or a PQC product: Quantum-Safe Certified, Quantum-Grade Security, Quantum-Safe vs Quantum-Resistant vs Post-Quantum and Patented Cryptography.
• Evaluating quantum key distribution (QKD) or a quantum random number generator (QRNG): Unhackable Quantum Encryption, Information-Theoretic Security and Quantum-Enhanced.
• Assessing a quantum computing company or its latest announcement: Post-Selection Laundering, Software-Defined Fault Tolerance and Circular Validation / Answer-Smuggling.
• Answering a board member or a vendor who says Q-Day has already arrived: RSA Is Already Broken and Quantum Apocalypse.
• Reviewing a cipher sold as perfectly or unconditionally secure: Perfect Secrecy, Operational Perfect Secrecy and Deterministic Chaos Encryption.
• Screening a retail investment offer: Quantum AI Trading and Quantum Financial System.

The Dictionary’s Two Tracks

I sort the terms into two tracks, because not all misleading terminology is equally dishonest. Of the 35 entries, 21 are Red Flag Terms and 14 are Misused Terms.

The test is whether a term has an established technical meaning. A term with none goes in the Red Flag track. A term with a real meaning that vendors stretch goes in the Misused track. Some cases are close. I classified zero-data PQC as a Misused Term because the preference behind it is sensible: keeping data out of a vendor’s cloud during migration. Vendors then sell that preference as post-quantum protection. Deterministic chaos encryption went to the same track because chaos theory is established physics and chaos-based cryptography is a legitimate research field, whatever vendors claim for it.

Four of the newer Red Flag entries name ways of presenting quantum computing results rather than labels on security products. Investors evaluating a quantum company need those four as much as CISOs evaluating a security product need the rest.

Red Flag Terms

Red Flag Terms have no established technical meaning in quantum physics, quantum computing or cryptography. Marketing departments invented some of them. Others join real words into phrases that mean nothing physically. A buyer who meets one in a pitch should ask for a technical definition, a standards reference, independent validation and a clear statement of scope. Some vendors who use these terms have real products and careless marketing, and a buyer learns which kind of vendor is in the room from the answers.

Proofs, Grades and Certifications That Do Not Exist

• Quantum-Proof: No post-quantum algorithm has a mathematical proof of security against quantum attack, which is why NIST, ETSI and IETF RFC 9794 all avoid the word. On its own I treat it as a yellow flag. When the vendor also declines to name its algorithm, it turns red.
• Operational Perfect Secrecy: Vendors keep the prestige of Shannon’s perfect secrecy and relax the condition that defines it, a key at least as long as the message. When they replace that key with a shorter secret, such as a ratchet state or a registration token, they bring back the computational assumptions that perfect secrecy was supposed to avoid.
• Quantum-Safe Certified: No organization anywhere issues a general quantum-safe certification. Under FIPS 140-3, accredited laboratories test specific implementations of approved algorithms, and a vendor claiming “quantum-safe certified” for a proprietary algorithm is claiming a credential that does not exist.
• Quantum-Grade Encryption: The quantum successor to “military-grade encryption.” No standards body defines it, and a vendor who uses it has named no algorithm, key size or security level.
• Quantum-Grade Security: “Quantum-grade security,” “quantum-level randomness” and their relatives read like certification tiers. The real tiers come from NIST SP 800-90B, FIPS 140-3, BSI AIS 31 and Common Criteria, and none of them has a quantum grade.
• Military-Grade Quantum Encryption: Neither “military-grade” nor “quantum encryption” applied to a classical product has a technical definition. Stacked together, the two halves sound twice as authoritative and add no information.
• Future-Proof Encryption: DES, MD5 and SHA-1 were each trusted for years before they were retired, and cryptographers do not describe even the new NIST post-quantum standards as future-proof. They aim for crypto-agility instead, the ability to replace an algorithm when it weakens.

Quantum and AI Vocabulary on Classical Products

• Quantum-Inspired Encryption: “Quantum-inspired” has a legitimate meaning in optimization and machine learning, where researchers adapt techniques from quantum computing for classical algorithms. A classical cipher gains no quantum security property by borrowing the vocabulary.
• Quantum Blockchain: A classical blockchain with a quantum random number generator or a post-quantum signature scheme added to it. Replacing ECDSA with a NIST-standardized post-quantum signature makes a chain post-quantum, which is useful, and leaves its computation as classical as before.
• Simulated Quantum Entanglement: Bell’s theorem shows that no local classical system can reproduce the correlations of entangled particles. A classical chip can compute the statistics, but a simulation has none of the security properties that physical entanglement provides.
• Wave Encryption: The phrase describes how ciphertext travels, by radio, light or sound, and says nothing about how it is protected. The cipher and the key protect a message before it enters any channel.
• AI-Powered Encryption: AI has a real and growing role in cryptography on the attack side, where researchers use it to break and stress-test ciphers. Vendors who claim that a model generates or certifies unbreakable encryption have the direction reversed.

Absolute Promises and False Alarms

• Removes the Attack Surface: After encryption, attackers go after key management, the implementation, the endpoints where data must be decrypted and the people with access. A product with no attack surface would also have no function.
• RSA Is Already Broken: The claim is false. RSA-2048 has not been broken, and the largest number factored with Shor’s algorithm on real quantum hardware is 21. Vendors use the claim to turn a future risk into a present emergency, and I examine the secret-break version in No One Has Secretly Broken RSA-2048 or RSA-4096.

Quantum Computing Results That Show Less Than Claimed

• Software-Defined Fault Tolerance: Vendors claim that software can remove the qubit overhead of quantum error correction. That overhead follows from the mathematics of fault tolerance. Error mitigation is real and useful, but it does not let a processor run circuits of arbitrary depth, which is what fault tolerance means.
• Post-Selection Laundering: Running a computation many times and keeping only the runs that pass quality checks is a legitimate research technique. It turns into laundering when a vendor presents the surviving runs as “zero logical errors” or as fault tolerance. The first question is how many runs the vendor threw away.
• Circular Validation / Answer-Smuggling: When a circuit’s parameters come from a classical solution of the same problem, the quantum computer confirms an answer that was already known. The test is whether anyone could have obtained the same result without the quantum hardware.
• Engineered Signature as Evidence: A vendor designs a system to drive every input to the same output, then presents that output as proof that the system works. Identical results across chips show only that the post-processing does what it was built to do.

Investment Scams and Pseudoscience

• Quantum AI Trading: Automated trading platforms that promise cryptocurrency profits from quantum computing. Financial regulators in more than a dozen countries have identified them as fraud, and no quantum technology is involved.
• Quantum Financial System: A conspiracy theory from the NESARA/GESARA movement that scammers use to sell fraudulent tokens, fake “activation fees” and pump-and-dump schemes. It has no connection to quantum technology.
• Quantum Consciousness: Princeton’s PEAR lab spent 28 years, from 1979 to 2007, testing whether human intention could bias random number generators. The lab closed, and a company founded by its alumni still sells consumer devices for “consciousness research.”

Misused Terms

Misused Terms come from physics, information theory and cryptography, where they have precise meanings established over decades of research. In marketing copy, vendors strip away the qualifying assumptions and drop the limitations. What remains implies far more than the physics or the mathematics supports.

Security Guarantees Stripped of Their Conditions

• Unconditionally Secure: A precise term for idealized protocols under stated assumptions. Vendors drop the assumptions and imply that the physical box in your rack is unconditionally secure, which no physical box is.
• Information-Theoretic Security: Security that does not depend on the attacker’s computing power, quantum computers included. The guarantee applies to a protocol under a model, and vendors present it as a property of their hardware and software.
• Perfect Secrecy: Shannon’s 1949 theorem requires a key at least as long as the message, used only once. A vendor claiming perfect secrecy with short or reused keys is contradicting a mathematical proof. The Shannon Hustle works through the mathematics and gives a five-question test for any such claim.
• Unhackable Quantum Encryption: QKD protocols have information-theoretic security proofs. Researchers have demonstrated detector-blinding, Trojan-horse and other side-channel attacks against commercial QKD systems, and the NSA, GCHQ and that published research all say the hardware is not unhackable.

Real Mathematics Sold as Security

• Undecidable Encryption: Hilbert’s tenth problem is undecidable, and ciphers built on it have been sold as unbreakable for three decades. Attackers keep breaking them because a cipher uses specific instances of the problem, and those instances can be far easier than the general case.
• Deterministic Chaos Encryption: Chaos theory is established physics, and chaos-based cryptography is a legitimate research field. Published critiques of chaotic ciphers go back to 1989, and NIST warns that passing statistical randomness tests is no substitute for cryptanalysis. However it is marketed, a deterministic chaotic cipher remains a symmetric cipher under a secret key.
• Unbounded Key Length: Past the standard sizes, a longer key stops adding security, because the key is no longer the weakest point. A 256-bit symmetric key already puts brute-force search out of reach, Grover’s algorithm included, so “infinite” and “million-bit” keys advertise a number without explaining where the security comes from.

Real Labels Applied Too Broadly

• Quantum Encryption / Quantum Cryptography: Both terms properly refer to QKD and related protocols whose security comes from quantum states. Vendors apply them to NIST’s post-quantum algorithms and to classical software with no quantum hardware at all, and I separate the three meanings in the entry.
• Quantum-Safe vs Quantum-Resistant vs Post-Quantum: Vendors use the three terms interchangeably, while NIST, ETSI, the NCSC and IETF RFC 9794 have each tried to pin them down. My own terminology guide covers the distinctions in depth. In this entry I focus on how sales teams exploit the overlap.
• Quantum-Enhanced: The broadest quantum-washing prefix. Vendors apply it to phones with real QRNG chips, which overstates a real component, and to classical software with no quantum hardware whatsoever, which is fabrication.
• Patented Cryptography: A patent shows that an idea is new and non-obvious. Patent examiners do not test security, so “patented” is commercial information, and real assurance comes from open analysis.

Sound Practices and a Real Threat, Oversold

• Air-Gapped Encryption: Air gaps and encryption are both real, and they answer different threats. If data never leaves the isolated network, post-quantum encryption in transit has nothing to protect there. If data does leave, the network is not air-gapped.
• Zero-Data PQC: Keeping data out of a vendor’s cloud during migration is a reasonable preference. Post-quantum protection comes from replacing the algorithms that protect data in transit, and vendors sell the first as if it delivered the second.
• Quantum Apocalypse: The threat behind the phrase is real: a capable quantum computer running Shor’s algorithm would break RSA and elliptic-curve cryptography. Vendors who say “apocalypse” wrap that threat in language built to sell panic. NIST has already standardized the defenses, and Q-Day is a migration problem with deadlines.

Companion Guide and Related Essays

I have written six longer pieces to go with the entries.

• How Quantum Snake Oil Vendors Respond When You Ask Hard Questions: The 16 most common deflection tactics, from “you’re funded by Big Tech” to pay-to-play magazine covers and predatory journals, with what legitimate companies say instead and the one question that reveals everything.
• Quantum Washing: The Quantum Industry’s Credibility Tax: The umbrella term for everything in this dictionary. Greenwashing had a decade-long head start and its own SEC enforcement playbook, and I trace how quantum washing is repeating that arc faster, because outsiders find quantum physics harder to evaluate than sustainability claims.
• The Shannon Hustle: The mathematics behind the entries on information-theoretic security, perfect secrecy, unconditional security and operational perfect secrecy, with a five-question framework for any claim of perfect secrecy at practical key sizes.
• You Cannot Detect Harvest Now, Decrypt Later (HNDL): Vendors have started selling HNDL detection, and boards ask for proof of HNDL targeting. Passive collection leaves no observable event inside a network, so neither is possible, and encryption is the defense.
• Quantum Snake Oil Is Flooding the News Cycle: Three extraordinary quantum claims in two weeks of June 2026, none peer-reviewed and all tied to commercial events, and what that pattern costs enterprises trying to make real migration decisions.
• Quantum AI Data Centers in Space Are an Absurdity: An investor asked me to evaluate a pitch for quantum AI data centers in orbit. The physics, the engineering and the economics all gave the same answer, which was no.

The Patterns Behind the Terms

Across the 35 entries and the essays around them, the same ten moves keep appearing. A buyer who recognizes them can evaluate a term that is not in the dictionary yet.

The three-card conflation. Vendors blur three different things. The first is physics-based quantum cryptography, mainly QKD. The second is post-quantum cryptography, classical mathematics designed to resist Shor’s algorithm. The third is a classical product with “quantum” added to the brand. A vendor says “quantum encryption,” and the listener cannot tell which of the three is on offer. The ambiguity is rarely accidental.

Theoretical-result laundering. A vendor cites a real result and strips away every assumption that makes it valid: Shannon’s perfect secrecy theorem, the information-theoretic security of the BB84 protocol, the undecidability of Hilbert’s tenth problem, the sensitivity of chaotic systems to their starting conditions, NIST’s standardization process. Each result is correct in its original setting, which is why it sounds unimpeachable in a pitch. The fraud is in the transplant.

Invented grades and certifications. Buyers are used to tiered assurance: four security levels in FIPS 140-3, seven evaluation assurance levels in Common Criteria, Type I and Type II SOC reports. Vendors copy the shape of those tiers with “quantum-grade,” “military-grade” and “quantum-safe certified,” and no issuing body backs any of them. Vendors present patents and cracking contests the same way, as if either were a security evaluation. The check takes one sentence: ask for the certificate number.

Absolutes in place of measurements. Vendors use “unhackable,” “unbreakable,” “future-proof,” “removes the attack surface,” “zero errors” and “infinite keys” to swap a measurable, bounded claim for an absolute one. Security engineers measure what remains exposed and explain how they contain it. A vendor who skips the measurement and promises elimination has given the buyer nothing to verify.

Selling one layer as protection for another. “Wave encryption” describes the transmission channel and leaves out the cipher. “Air-gapped encryption” mixes up network isolation with protection for data in transit, and “zero-data PQC” mixes up where data moves during migration with how it is protected afterward. A QRNG improves one input to key generation, and “quantum-enhanced” credits it with upgrading the whole system. In each case the vendor is right about one layer and sells it as protection for another.

Demonstrations that cannot fail. In each of the four quantum computing entries, the demonstration could not have produced a different answer. Post-selection keeps only the runs that passed. Circular validation loads the known answer into the circuit. An engineered signature forces every chip to the same output. Vendors of software-defined fault tolerance claim to remove an overhead that the mathematics of error correction requires. An investor learns nothing about a quantum computer from a result that could not have come out differently. A buyer can expose the pattern with four questions: how many physical and how many logical qubits, error detection or error correction, how many runs were discarded, and whether anyone knew the answer before the run.

Proprietary algorithm secrecy. Kerckhoffs’s principle, published in 1883, requires that a cipher system stay secure even if it falls into enemy hands, so that only the key has to stay secret. A vendor who says its algorithm must stay secret is breaking a foundational rule of the field. If the algorithm has not been through public standardization or published cryptanalysis, and the vendor cannot name it, walk away. Proprietary quantum-safe algorithms remain the single most reliable indicator of snake oil in this market, a point I have made in Quantum of Flapdoodle, in Q-FUD and in my analysis of proprietary post-quantum algorithms.

Credibility laundering. Pay-to-play magazine features, trade show awards given to every exhibitor who applies, papers published in predatory journals, partnerships that are actually customer relationships and credentials from unaccredited institutions. The companion guide documents 16 variants of this pattern.

Q-FUD timeline arbitrage. Vendors move Q-Day closer in their marketing, all the way to the present in the “RSA is already broken” pitch, and sell the distance between that date and the buyer’s migration plan. The real urgency argument has nothing to do with whether Q-Day arrives in 2030 or 2035. The deadlines are already set by regulators, insurers and supply chain requirements, regardless of when a CRQC arrives.

Denial as grift. Denial is the mirror image of Q-FUD. Instead of moving Q-Day closer to sell products, denial grifters push it to infinity so that nobody has to pay for migration. I used Bitcoin as a case study: at Bitcoin 2026, the organizers put engineers building quantum-resistant upgrades on the same main stage as three speakers who argued that quantum computers can never work because Bitcoin proves time is discrete. Organizations that dismiss the quantum threat and organizations that panic-buy snake oil end up in the same place, unprepared when the deadlines arrive.

For a lighter test of the same patterns, my Quantum Technobabble Generator writes a plausible quantum-flavored paragraph on demand and gives a verdict on any paragraph you paste into it. Try it on the next vendor press release you receive.

What Legitimate Vendors Say Instead

The most useful comparison for a buyer is with legitimate vendors, because their answers are specific enough to check.

A vendor building real post-quantum security names its algorithms: ML-KEM (formerly CRYSTALS-Kyber), ML-DSA (formerly CRYSTALS-Dilithium) and SLH-DSA (formerly SPHINCS+), which NIST published as FIPS 203, 204 and 205 in August 2024, or another algorithm with a comparable public record of cryptanalysis. It can quote the certificate number of a cryptographic module validated under FIPS 140-3, or say when it expects validation. A QRNG vendor can show an entropy source validated under NIST SP 800-90B, and my QRNG buyer’s guide explains why a buyer should weigh that certificate above the word “quantum” on the box.

A QKD vendor with a real product describes the attacks its hardware defends against, detector blinding included, and the attacks it leaves to other controls. A quantum computing company reports physical and logical qubits separately and says whether it demonstrated error detection or error correction. It also publishes how many runs it discarded.

Legitimate vendors welcome a buyer’s own cryptographer or physicist. Their work appears in recognized journals and standards processes, and a technical question gets a technical answer.

Ten Questions to Ask a Quantum Vendor

Most entries end with questions specific to one term. These ten work across the whole dictionary.

1. Which algorithm does the product use, by its standard name, and which standard specifies it?
2. What is the certificate number? For a cryptographic module that means a certificate from NIST’s Cryptographic Module Validation Program under FIPS 140-3, and for an entropy source an SP 800-90B validation.
3. What is the secret, how many bits of entropy does it contain, and how does the other side obtain it?
4. Where is the full specification, and who outside the company has published an attack on it?
5. Which threats does the product address, and which does it leave to other controls?
6. What does the quantum component change, compared with a classical product with the same certifications?
7. If RSA or elliptic-curve cryptography is already broken, as the vendor says, then broken by what method, on what hardware, and published where?
8. For a quantum computing result, how many physical and how many logical qubits were used, was it error detection or error correction, what fraction of runs was discarded, and was the answer known before the run?
9. Can I bring my own cryptographer or physicist to review the claim?
10. What result would show that the claim is false?

For a detailed guide to reading the answers, see the companion article on vendor deflection tactics. For a broader view of quantum hype, start with Quantum of Flapdoodle. For the Q-FUD problem specifically, see Q-FUD: The Quantum Panic Industry. For a case study in quantum washing, see China’s “Photonic Quantum Chip” Is Impressive. But It’s Also a Case Study in Quantum-Washing.

The threat behind all this marketing is real, and the PQC migration has to happen regardless. For the practical response, see Practical Steps to Quantum Readiness, the PQC Migration Framework and my book Quantum Ready.

A Note on Intent

Most entries open with a note on scope, and the capstone needs one too. I am not accusing any specific company or individual of fraud. These articles examine terms and claims on their technical merits. A company using misleading terminology might be running a scam, or it might be a legitimate technology company with an overeager marketing department. The physics is the same either way, and so are the questions I suggest asking.

Technology also evolves. A term I flag today may have a clearly defined, defensible technical meaning in five years, and if that happens I will update the entry. In the deterministic chaos entry, for example, I committed to revising the verdict if a product sold under that label publishes a full specification and withstands years of open cryptanalysis.

The physics will not change. Under Bell’s theorem, no local classical model will reproduce the correlations of entangled particles. Under Shannon’s 1949 theorem, perfect secrecy will still require a key with at least as much entropy as the message. A deterministic generator with finite memory and no fresh input will still fall into a repeating cycle, as the pigeonhole principle requires. Under Kerckhoffs’s principle, now more than 140 years old, security will still have to depend on the key alone. In every entry I measure a marketing claim against fixed points like these.

Right of Reply and New Terms

I welcome corrections and substantive right-of-reply requests. If a standards body, regulator, vendor, researcher or reader can identify a recognized technical definition, certification scheme, independent validation or relevant correction for any entry, I will review it and update the article where appropriate.

If you meet a quantum term in a sales pitch that is not in the dictionary yet, send it to me with the context in which you heard it. The simulated quantum entanglement entry started the same way, with readers and colleagues who had met the phrase in vendor pitches.

From the author