Zero white papers. Zero testnets. Zero named cryptographers. Zero audited code. That is the complete evidentiary record behind AmericanFortress, a project claiming it can shield existing Bitcoin, Ethereum, and Solana wallets from quantum computers without a single address change and without forcing a single user to migrate funds.
The announcement slid through the crypto press with all the friction of a press release strategically dropped before a holiday weekend. The market's silence has been deafening. In a narrative-driven cycle, a "quantum-safe, backward-compatible encryption scheme" attached to three of the largest asset chains in existence should have triggered at least a flicker in trading desks, a thread of wallet-provider debate, a wave of skeptical developer retorts. None of that happened.
Why? Because the professionals who actually build wallet infrastructure read one sentence — "no migration, no address changes" — and did the math. The math does not work in public. Not with today's signature schemes. Not with today's address formats. Not without a cryptographic breakthrough that would rewrite the rules of post-quantum research. A breakthrough, notably, that AmericanFortress has not documented.
I have spent the better part of two decades auditing infrastructure claims against code rather than narratives. The 0x protocol audit sprint in 2017 taught me that the fastest way to find the truth in crypto is to read the contract and ignore the press kit. The Terra-Luna collapse in 2022 taught me that on-chain data outsells official narratives every single time. In that spirit, I went looking for the AmericanFortress technical documentation. I found none. That is not a small detail. That is the story.
Why This Claim Collides With Cryptography
Start with the threat model, because it is real even if the timetable is disputed. Shor's algorithm, published in 1994, demonstrated that a sufficiently powerful quantum computer can factor large integers and compute discrete logarithms in polynomial time. That matters because the entire edifice of modern cryptocurrency — Bitcoin's ECDSA over Secp256k1, Ethereum's ECDSA signatures, Solana's Ed25519 — rests on the hardness of elliptic curve discrete logarithms. A quantum computer with enough logical qubits and sufficient coherence could, at least theoretically, derive private keys from public keys and forge signatures at scale.
The word that matters is "theoretically." As of 2025, no quantum machine has demonstrated the logical qubit count, error-correction fidelity, and circuit depth required to break Secp256k1 within a realistic attack window. Industry estimates for a cryptographically relevant quantum computer range from five to twenty years, depending on who is modeling the assumptions and whose grant is paying for the projection. Google's Willow chip in late 2024 produced headline-grabbing error-correction research; it did not produce the ability to crack ECDSA.
Still, the cryptographic community has not been idle. NIST ran a multi-year competition to standardize post-quantum cryptography. The winners — ML-KEM (formerly CRYSTALS-Kyber) for key encapsulation, ML-DSA (formerly CRYSTALS-Dilithium), Falcon, and SPHINCS+ for signatures — are already being rolled out across internet infrastructure. Standards get migrated. Certificates get rotated. Protocols get patched.
Blockchain is a different beast entirely. The problem is not the absence of quantum-safe signature algorithms. The problem is that those algorithms produce different public keys, different signature sizes, and different algebraic structures than the elliptic curve schemes baked into every wallet address in circulation today. A Bitcoin address is not the public key; it is a hash of the public key. The signature format is entangled with the scripting language. The address standard encodes assumptions about the underlying cryptographic primitive. Swap in Dilithium and you change the address derivation, the transaction weight, the block-space economics, and the consensus validation rules all at once.
Existing competitors in the quantum-safe blockchain space prove the point. Quantum Resistant Ledger ships a chain built natively on Lamport/Winternitz signatures, and it still requires users to move to a brand-new blockchain. QANplatform offers Ethereum Virtual Machine compatibility with Dilithium for newly generated addresses, but it does not retrofit legacy ETH addresses. Algorand has published research paths toward post-quantum signing, but the transition plan involves network-wide coordination. Even the most serious, well-funded projects in this vertical have never claimed that legacy addresses could survive a signature algorithm swap without visible change.
The commodification of quantum fear is its own market. A search across crypto press archives shows a cottage industry of "quantum-resistant" announcements over the past five years. Most follow the same arc: a concept announcement, a wave of generic coverage, a token sale or a security product launch, then silence. The pattern is so consistent that I have started treating "quantum-proof" in a headline as a risk flag rather than a technical descriptor. Real cryptographic progress does not need a PR push; it needs a paper, a proof, and a peer group willing to tear it apart.
That is why a claim of "no migration, no address changes" is so extraordinary. It implies either the discovery of a completely novel construction — something that would constitute a landmark in applied cryptography — or a fundamental misunderstanding of how existing blockchains operate. My read of the available information favors the latter.
The Address Is a Commitment. That's the Problem.
Let me lay out precisely why "quantum-safe without address changes" collides with the mathematics of live chains.
On Bitcoin, a standard P2PKH address is a Base58Check encoding of Hash160 — RIPEMD-160 over SHA-256 — of a Secp256k1 public key. On Ethereum, an address is the last 20 bytes of the Keccak-256 hash of the unpadded public key. On Solana, addresses are the base58 of 32-byte Ed25519 public keys with a trailing checksum. The common thread: the address is a commitment, a hash output, not the public key itself.
The industry has long interpreted this as a safety feature. A quantum attacker cannot reverse a hash commitment directly, so funds sitting in untouched addresses appear protected. The protection vanishes the moment a public key is exposed. On Bitcoin, public keys appear on-chain whenever a user spends from an address; the scriptSig reveals the raw public key. On Ethereum, every transaction exposes the sender's public key through ECDSA signature recovery. On Solana, the same pattern applies. Even if the address remains a hash, public key material leaks into public view with each spend. Once exposed, Shor's algorithm can mathematically extract the private key at sufficient quantum scale.
Researchers have flagged this "address reuse, revealed key" problem for nearly a decade. Post-quantum signature schemes exist specifically to close that gap on future transactions. But no one has seriously proposed retrofitting existing legacy addresses, because addresses are derivations of elliptic-curve public keys, and the signature scheme is entangled with every transaction's validity. You cannot swap the signature algorithm without forking the chain, changing the script protocol, or altering address formats. These are not cosmetic details. They are consensus rules.
AmericanFortress claims to have solved all of this without touching any of it. That requires a mechanism that validates a post-quantum signature against a commitment that remains a hash of an elliptic-curve public key, without ever revealing or depending on the elliptic-curve public key as the source of truth. That is not a modest innovation. It is a new category of signature scheme that no cryptographer has published, no IETF draft has proposed, and no peer-reviewed venue has examined.
I have seen theoretical attempts in this direction, mostly involving zero-knowledge proofs and novel verifiable random function constructs. Zero-knowledge mechanisms can, in principle, prove knowledge of a preimage to a hash without revealing it. Extending that to post-quantum key derivation is an active research frontier, not a production capability. Any such system would drastically increase computational requirements, signature verification costs, and transaction data sizes. The AmericanFortress announcement contains no indication that those engineering constraints have been addressed.
Some may argue that "quantum-safe encryption" could refer to a client-side encryption layer — encrypting keys or mnemonic seeds with a lattice-based scheme while leaving the underlying addresses untouched. That interpretation collapses under the same public-key exposure problem. Encrypting a private key at rest does nothing if the public key derived from it is broadcast during a transaction. The quantum attack surface is the public key in the mempool and in the transaction history. Local encryption is irrelevant to that exposure. If the AmericanFortress team is pursuing this angle, they have misunderstood the threat model on a basic level.
Chain by Chain: The Specifics
Let me go chain by chain, because the details matter and each chain blocks this claim differently.
Bitcoin. ECDSA signatures on Secp256k1 run 71 to 73 bytes. Falcon-512 signatures run 666 bytes. Dilithium runs over 2.4 kilobytes at its primary security level. SPHINCS+ runs even larger. Bitcoin's block weight limit is four million units. Even a handful of post-quantum-signature transactions would consume an order of magnitude more space than an equivalent ECDSA bundle. Taproot, which introduced Schnorr signatures as a conservative upgrade that preserved address compatibility, took years to reach majority adoption. A full PQ signature migration implies a deeper script-level rewrite, a new address variant carrying new key types, and consensus changes touching every node and wallet on the network.
Ethereum. ECDSA signatures are 65 bytes fixed. The Ethereum Virtual Machine validates signatures in precompiled contracts. Replacing that with a PQ scheme means new precompiles, new gas schedules, new transaction types, and a coordinated hard fork across the validator set. Account abstraction (ERC-4337) offers a partial path by allowing contract-based accounts to define custom validation logic. But existing externally owned accounts, the hash-derived addresses that dominate the chain, would still need migration or a new account type. The "no migration" claim runs directly into this wall.
Solana. Ed25519 signatures are 64 bytes. Solana's throughput model depends on lightweight signature verification across tens of thousands of transactions per second. Dilithium verification is orders of magnitude slower in current benchmarks, and its payloads are dramatically heavier. Solana would need a full validator upgrade, a new transaction format, and ledger replay considerations. Again, no path preserves the current address set without coordinated network changes.
I am not claiming that a layered solution is impossible. A hardware wallet firmware update, a custodian-side service, or a secure-enclave helper could add a post-quantum amplification layer on top of existing infrastructure. But those solutions protect specific endpoints. They do not touch the consensus layer. An adversary with a cryptographically relevant quantum computer targets public keys on-chain and signatures in the mempool, not a wallet app's local encryption. An infrastructure-layer "protection" that does not alter protocol validation is security theater painted over a cracked hull.
The Performance Dead End
Suppose AmericanFortress did have a post-quantum construction embedded in a device or enclave. The next question is throughput. Post-quantum signatures are not only large; they are computationally expensive to verify. Dilithium verification is roughly an order of magnitude slower than ECDSA in typical benchmarks. Falcon is faster but still substantially larger. On a chain like Solana, validator performance would degrade unacceptably. On Ethereum, calldata costs alone would make PQ-signed transactions prohibitively expensive at realistic gas limits. On Bitcoin, the weight limit constrains how many PQ transactions fit in a block — potentially dozens instead of thousands.
None of these constraints appear in the AmericanFortress announcement. No performance data. No benchmarks. No testnet. For a scheme claiming universal applicability to three major chains, the absence of a single measurement of transaction size, verification cost, and chain-level throughput impact is disqualifying on its own.
What a Real Post-Quantum Migration Looks Like
For readers not spending nights with NIST documents, here is what an honest migration actually looks like.
Phase one is research and standardization: a consortium of cryptographers agrees on algorithms, security parameters, and attack models. That work is done. ML-DSA and ML-KEM are published. Phase two is protocol design: blockchains define new address formats, new signature validation rules, and new wallet derivation paths. That is underway in a few ecosystems, but nowhere near consensus. Phase three is ecosystem rollout: every wallet, exchange, custody provider, hardware manufacturer, and protocol must release compatible updates. Phase four is migration education: users and businesses must understand how to move assets and rotate keys safely. Phase five is the transition period itself: old and new systems must interoperate securely for years, possibly more than a decade.
That timeline is not bureaucratic caution. It is the pattern of every major cryptographic transition in history, from the migration of TLS away from RSA to the SHA-1 deprecation. Each required years of public warnings, dual-run periods, and sunset dates. Crypto's decentralized consensus model makes it slower, because every node, validator, and wallet operator must choose to upgrade voluntarily. Bitcoin took years just to reach Taproot, which merely added a more flexible script without touching the foundational curve. A post-quantum hard fork attending to address format changes, wallet updates, exchange integration, and hardware firmware rollout makes Taproot look like a weekend patch.
The most serious post-quantum claims in the industry — from Algorand's research team, from the IETF's working groups, from academic labs — all follow that arc. None skip the white paper. None skip peer review. None skip the testnet. None claim unilateral backwards compatibility with legacy address schemes.
AmericanFortress skipped every phase. That does not mean the company is fraudulent. It means the claim is not experimentally testable, and a claim that cannot be tested does not deserve to be taken seriously.
The Six-Evidence Test
I have a framework I apply to claims like this. I first developed the discipline in 2017 while reverse-engineering the 0x protocol v2 fillOrder function during a 72-hour audit sprint. The pull request I submitted after finding the reentrancy path was merged. That experience set a permanent rule: never treat a project's own words as the unit of analysis. Look at the artifacts.
AmericanFortress fails a six-point evidence test in public.
One: a public white paper naming cryptographic algorithms. Absent. No lattice construction is named. No key encapsulation mechanism is described. No post-quantum signature family is cited. "Quantum-safe encryption scheme" appears as a claim, not a specification.
Two: a reference implementation on a testnet. Absent. A claim of protocol-level compatibility with Bitcoin, Ethereum, or Solana requires, by definition, a live testnet showing transactions signed and verified. There is no testnet.
Three: a third-party security audit. Absent. No Trail of Bits report. No Quantstamp review. No NCC Group engagement. No Consensys Diligence assessment. A cryptographic claim of this magnitude without an independent audit is an undeliverable all by itself.
Four: named cryptographers with verifiable track records. Absent. The strongest signal of serious cryptographic work is the willingness of an established researcher to attach their identity to the claim. The Ethereum Foundation's quantum research group publishes named authors on every paper. NIST standards carry named contributors. AmericanFortress appears to carry nobody.
Five: compatibility testing with existing wallet software and address formats. Absent. A claim of "no address changes" should come with a demonstration that a transaction from a legacy address verifies under the new scheme. No such demonstration exists.
Six: consensus on the roadmap. Absent. No wallet provider, exchange, or standards body has endorsed the approach. No open-source community has reviewed the code. No public repository has been linked.
Six items. Zero hits. This evidence vacuum is itself data. The Terra-Luna experience is instructive. When Anchor Protocol's withdrawal queue started growing in May 2022, I did not wait for official reports. I pulled transaction flows directly from explorers. The data showed a cluster of whale wallets that had been exiting for 48 hours before the public narrative shifted. The chain told the truth before anyone issued a statement. The truth about AmericanFortress is similarly available in what has not shipped: no on-chain activity, no repository activity, no auditable footprint.
I deployed the same instinct two years earlier, during the 2020 DeFi Summer. Gas prices on Ethereum spiked before mainstream coverage of the first flash-loan attacks arrived. I tracked transactions to the Uniswap V2 pairs, spotted the draining pattern, and posted an alert minutes after the first anomaly. The principle holds: the infrastructure never hides as well as the marketing does. AmericanFortress's infrastructure is not hiding. It is nonexistent.
What the Audit Experience Says
The most dangerous claims in crypto are not the ones that are obviously fraudulent. They are the ones that wrap a kernel of truth in a narrative of urgency. I learned that lesson again in 2024 while auditing the public filings of three asset managers ahead of the Bitcoin ETF approval. Their marketing materials boasted multi-signature key custody. Their registered filings told a different story: a single trust company, a single hardware module, a single vendor dependency for the entire custody stack. The gap between the disclosure and the narrative was not a random error. It was the system working as designed.
The same pattern repeats here. AmericanFortress's pitch wraps a real problem — the eventual arrival of cryptographically relevant quantum machines — in an impossible promise. The promise is designed to be accepted on faith, because the underlying math cannot be examined in the absence of a white paper. Fear does the heavy lifting. Verification never gets a chance to load.
My rule, refined over years of auditing both code and filings, is simple: examine the gap between what a project claims and what it ships. For AmericanFortress, the gap is the entire project. The claim says "quantum-safe, no migration." The reality says "no specification, no code, no testnet." That gap is not a minor deficiency. It is the data point that matters most.
There is one more consideration that emerges from my 2024 audit work: custody and key management are where security claims most often break. Even a perfect post-quantum address scheme collapses if the private key is stored on a single compromised device. Post-quantum cryptography protects against Shor's algorithm. It does not protect against a poorly managed custody stack, a phishing email, or an insider. The AmericanFortress announcement does not address key management at all. That silence is not neutral; it reveals the level of rigor behind the claim.
The Scam Playbook
This brings me to a darker consideration. The crypto industry has a documented history of leveraging quantum fear as a social engineering vector.
The playbook is straightforward. Step one: announce a "revolutionary" quantum-resistance solution to an existential threat. Step two: keep technical details vague, citing intellectual property concerns. Step three: generate press coverage through a thinly sourced release rather than a conference paper. Step four: if attention accumulates, introduce a token, an investment round, or an API product that requires users to deposit funds, grant approvals, or install software. Step five: vanish.
I am not asserting AmericanFortress is fraudulent. There is insufficient data to prove intent. But the pattern is familiar, and the absence of technical disclosure is not a neutral omission. In cryptographic practice, secrecy functions as a warning. Legitimate projects publish early precisely because the field advances through open scrutiny.
There is also a narrower practical risk: a "no migration" message may reassure users who are anxious about quantum threats into a false sense of security, steering them toward software that claims to protect them. A project with no verifiable security posture should never be the foundation of anyone's wallet safety strategy. At best, it is aspirational research. At worst, it is a phishing lure wearing a lab coat.
The strategic absurdity is worth emphasizing. If AmericanFortress genuinely possessed a backward-compatible, quantum-safe construction that preserved Bitcoin, Ethereum, and Solana addresses, it would be the most valuable cryptographic patent in history. Every major exchange, custody provider, and wallet vendor would enter a bidding war within days. The rational move would be to publish a proof of concept and license the technology. Publishing a vague announcement without protecting the intellectual property through a white paper makes no commercial or engineering sense. The rational inference is that there is no proof to publish.
Source-level verification, meanwhile, remains the industry's weak point. I have spent years at the editor's desk watching major stories — Terra, FTX, the DAO hack — pass through newsrooms with minimal technical validation. The pattern repeats because the cadence rewards speed, and speed rewards whoever repeats the loudest claim first. That is why I built my team around forensic verification: every market-moving story gets an on-chain transaction check, a contract check, or a document check before we run it. AmericanFortress would not have survived that filter. It barely survives this article.
The Contrarian Blind Spot
Now let me push against my own skepticism, because this industry deserves friction from both directions.
The quantum threat is overhyped on a one-to-three-year horizon. I hold that view. But the comfortable industry timeline — ten to twenty years until a cryptographically relevant machine — is only comforting if the migration clock starts now. It is not starting. The full migration of a chain-wide signature scheme is itself a decade-long project. Draw the two curves on the same chart and the gap narrows uncomfortably. A machine in twenty years against a migration that starts in ten and finishes in ten after that leaves a hard zero margin for error. The infrastructure refuses to move faster because the industry keeps treating quantum migration as a distant hypothetical. That indifference is a real blind spot, and projects like AmericanFortress are what fill blind spots with noise.
The second nuance concerns asset risk classification. The highest-risk coins are not legacy ECDSA addresses. They are the ancient P2PK Bitcoin UTXOs from 2009-2011 that expose public keys directly on-chain. They are reused addresses in cold storage whose public keys sit on block explorers. They are multisignature custody schemes and cross-chain bridge keys. They are exchange hot wallets managed in software. A scheme that preserves existing addresses does nothing for these exposed pools. For the assets that need protection most, the answer requires movement into new key structures. "No migration" is the wrong slogan for exactly the assets the claim pretends to save.
The phrase is a UX dream and a security nightmare bundled into one. The dream is real: every wallet provider and custodian dreads a mandatory migration that would strain retention and create operational chaos. Whoever eventually solves a painless post-quantum migration owns a foundational layer of the crypto economy. But the solution will come from rigorous standards work, named researchers, exchange consortiums, and years of interoperability testing. Every serious infrastructure shift in this industry — SegWit, Taproot, EIP-1559 — succeeded through patient cross-industry coordination, not a unilateral breakthrough decree from an unknown entity.
I ran a similar test on the NFT market in early 2021, when I audited the metadata JSON files of a trending PFP collection. The floor price looked healthy; the actual availability of the images was not. Fifteen percent of the collection's metadata lived on centralized IPFS gateways that were silently failing. The assets were, in practical terms, partially invisible. The lesson generalizes: the surface narrative and the underlying infrastructure are often two different stories. Quantum-safe claims deserve the same treatment. Check the gateway. Check the artifacts. Ignore the floor price.
Consider the asymmetry of error. If the industry overreacts to the quantum threat and migrates signature schemes earlier than strictly necessary, the cost is measurable: engineering hours, user friction, transition bugs. If the industry underreacts, the cost is existential: a mature quantum machine arrives, public key material from the last fifteen years is compromised, and entire asset classes become unrecoverable. That asymmetry argues for starting the migration now, even if the immediate threat is years away. It does not argue for accepting the first "quantum-proof" solution that appears with no evidence. The right response to the threat is disciplined engineering, not desperate gullibility.
The event of real quantum migration will not arrive through a press release. It will arrive as a set of NIST references updated in a wallet's BIP draft, a testnet upgrade in a client repository, an exchange's announcement of a new address format with a sunset schedule for the old one. Those moves are unglamorous, incremental, and verifiable. They are also the only moves that matter.
Let me also flag the regulatory dimension. There is no token here. No securities offering has been described. That keeps the current regulatory posture simple. But if AmericanFortress later attaches a token or a paid API service to this narrative, the risk surface expands dramatically. A "quantum-safe" token sale would combine novelty and fear, the two emotional ingredients most reliably harvested by crypto fraud. Watch for it.
Takeaway: The Watchlist
Here is what I will be watching over the next six months, and what any serious analyst should watch as well.
First, a white paper naming a specific algorithm family. The presence of a NIST-standard algorithm like ML-DSA, Falcon, or a concrete lattice-based construction would at least make the claim scientifically falsifiable. Absence of named algorithms is dispositive.
Second, a testnet demonstrating the scheme. Transactions signed. Addresses preserved. Verification working under realistic load. A testnet is not optional; it is the minimum unit of cryptographic credibility.
Third, a named cryptographer with a verifiable publication record. The willingness of an established researcher to attach their identity to the claim is the strongest signal of intent. Its absence signals the opposite.
Until those three artifacts manifest, the rational posture is neither quantum panic nor quantum peace. It is a cold-eyed audit of each claim as it crosses the desk. Chaos is just data waiting to be organized. The data available here is already organized: no code, no testnet, no team. The chain will not lie about that.
What you see on-chain is not always what you get. In this case, there is nothing on-chain at all. That is not a bug in the analysis. That is the analysis.
Security is a promise; liquidity is the proof. AmericanFortress has made a promise without proof. The market delivered its verdict in silence. Volatility is not the market's exception; it is the market's immune response to unverified narratives. Watch the immune system do its work.
The market's indifference is the correct verdict. But indifference is not the same as safety. The quantum clock is running. The industry's job is to build the migration on the right timescale, with the right standards, and the right evidence thresholds. AmericanFortress demonstrates what happens when the demand for a solution meets the absence of a real one: a vacuum, filled by noise. The answer is not to ignore the vacuum. It is to hold every claimant to the same standard: show the code, name the researcher, run the testnet.
When the next quantum-safe claim hits the wire, ask only one question: whose name is on the paper, and where is the code? If the answer is silence, keep your funds in the chain you actually trust. The quantum migration will arrive. The real builders will bring testnets, not teasers.