🔍 Read the full analysis: A Changing Security Landscape: AI, Quantum Computers And Encryption on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
OpenAI published 722 AI-generated mathematical manuscripts on Oct. 6, including results involving computational complexity. The release has prompted questions about whether AI could find algorithms that weaken cryptographic systems, including post-quantum standards, but no cryptographic break has been reported and the mathematical claims remain subject to verification.
OpenAI published 722 mathematical manuscripts on Oct. 6, generated by an unreleased internal model, prompting debate about whether AI could find new algorithms that weaken encryption. No cryptographic system has been shown to be broken, and the manuscripts’ claims require verification, but the discussion reaches beyond the better-known threat from future quantum computers to the mathematical assumptions behind both current and planned security systems.
The manuscripts covered 372 families of problems and were selected from work on roughly 4,000 problems, according to the source account. Reported results include claims about integer multiplication and Fourier transforms running below previously expected time bounds. A separate result published the day before by Virginia Vassilevska Williams and Josh Alman reportedly used an idea from an Anthropic model to improve the known time for 3SUM, a computational problem long associated with a near-quadratic bound. These developments concern algorithms, not a demonstrated attack on encryption.
Computer scientist Scott Aaronson drew attention to the absence of cryptography from OpenAI’s collection and said his sources indicated that AI companies were discreetly testing whether internal models could break important protocols. That testing has not produced a publicly reported cryptographic break in the supplied material. The distinction matters: a model finding a faster algorithm for a mathematical problem is not, by itself, evidence that it can recover encryption keys or defeat a deployed protocol.
The source also reports that OpenAI withdrew a claimed proof concerning the Hodge conjecture for products of K3 surfaces after a sign error was identified. That correction illustrates the need for independent checking. The manuscripts are claims requiring scrutiny, rather than settled mathematical results; the source does not establish that all 722 have been validated.
The old map is gone: AI mathematics, quantum computers and the cryptography holding up finance and defence
For a decade the plan was simple: elliptic curves doomed by quantum; lattices safe; hashes safe. Nothing has been broken. But a second threat has arrived that doesn’t respect those borders — AI producing new mathematics faster than any human community, against assumptions that are believed, not proven.
Now: on borrowed time — possibly shorter than the quantum countdown suggests.
Now: unproven against AI — and the destination most of the world is migrating to.
Now: reminded estimates move — BSI advised against new deployments on 1 Oct 2026.
Now: safest ground available — not a guarantee.
~n log0.9999999999999 n — a barrier many thought fundamental (OpenAI, claimed)
Overturns a half-century conjecture. Williams & Alman; key idea from an Anthropic model
“Conspicuous by its absence” (Aaronson) — labs reportedly testing crypto “gingerly and discreetly”
ECDSA could break before Q-day, “in the worst case in months not years.” Move funds to never-signed addresses. ~6M BTC sit behind exposed keys.
The new risk is the destination of the migration. Hash-only where possible; “much more paranoid” lattice params; ×10 key sizes long-term. Doesn’t recommend anyone scramble.
“No evidence whatsoever” that elliptic-curve assumptions are close to failing.
Classical breaks could reach “quantum-safe” schemes — but don’t treat a two-year scenario as a date.
Known to IBM and the NSA designing DES (~1974); public via Biham & Shamir (~1990); confirmed by Coppersmith (1994).
Invented at GCHQ — RSA- and Diffie–Hellman-equivalents — and kept secret for over two decades.
No crypto in 722 manuscripts. Found and withheld? Not posed? Posed and failed? Indistinguishable from outside.
Traffic recorded today is decrypted when a break arrives. For secrets that must last 25+ years, a break in 2035 is a break today. A state that finds one won’t announce it — it will mine its archives.
Signatures can be built from hashes. Encryption and key exchange need a trapdoor with structure — lattices, codes or group theory. Defence can only choose which structure, how much margin, how many combined.
Every date was set against quantum hardware forecasts with visible warning. The AI threat offers none.
“ML-KEM everywhere” means starting over if lattices weaken. “We can swap algorithms” doesn’t.
Blockchains show a classical break first — exposed keys and balances are public. Monitor dormant exposed addresses.
Every algorithm, key, certificate, protocol.
PQ + classical, as BSI requires.
Firmware, updates, long-term keys.
Highest sets; evaluate FrodoKEM.
More than one mathematical family; HQC coming.
Swap algorithms without rebuilding.
Forward secrecy, rotation, hidden keys.
Buterin: lost more in botched migrations than in all hacks.
Nothing has been broken, and the sceptics are right that there’s no evidence elliptic curves or lattices are about to fall. But the map has changed: elliptic curves on borrowed time, lattices unproven against AI, codes reminded that estimates move, hashes the safest ground available. For finance, intelligence and defence the answer is the same whichever threat arrives first.The quantum threat comes with a countdown. The AI threat may arrive as a silence — an empty folder where a paper should have been. The winners will be those who can change their algorithms fastest.
Why Cryptographic Planning Is Changing
Governments, banks and military organizations are already preparing for the possibility that sufficiently capable quantum computers could break widely used public-key systems. The standard response has been migration to post-quantum cryptography. The new concern is that AI might discover a classical algorithm that changes the security picture without requiring quantum hardware.
That possibility matters because cryptographic security depends on mathematical problems being difficult to solve, not on a proof that they are impossible to solve efficiently. If an algorithm substantially reduced the cost of attacking a system, organizations might need to revise security assumptions, replace protocols or increase key sizes. But that is a risk scenario, not a reported outcome. The source gives no evidence that current AI systems have found such an algorithm for a deployed cryptographic standard.
The warning signs could also differ. Quantum-computing progress can be followed through hardware development and engineering milestones. A useful algorithm could, in principle, be discovered and kept private. That prospect complicates risk planning for finance, intelligence and defence, although the supplied material does not establish that any government or company has achieved a secret cryptographic break.
As an affiliate, we earn on qualifying purchases.
Quantum Migration Meets AI Research
The existing quantum concern centers on Shor’s algorithm, which could threaten RSA and elliptic-curve cryptography if run on a sufficiently large, error-corrected quantum computer. In August 2024, the U.S. National Institute of Standards and Technology standardized post-quantum alternatives: ML-KEM for establishing encryption keys, ML-DSA for digital signatures, and SLH-DSA, a hash-based signature standard. Organizations have been planning migrations toward these and other post-quantum tools.
AI-generated mathematics raises a different question: whether new classical algorithms could weaken assumptions behind existing systems or the newer lattice-based standards. The source does not show that this has happened. It reports, instead, that AI systems are producing mathematical results that challenge some expectations about computational speed, while cryptographic security itself remains unbroken in the account.
On Oct. 7, Ethereum Foundation researcher Justin Drake urged the crypto industry to plan calmly for a possible “bunker mode,” including moving funds to addresses whose public keys have not been exposed. He warned that elliptic-curve signatures might be at risk sooner than the anticipated quantum timeline, but framed this as a possibility. Ethereum co-founder Vitalik Buterin cautioned against rushing to move funds and directed attention to the security assumptions of lattice-based systems and other emerging technologies.
“calmly begin planning for ‘bunker mode'”
— Justin Drake, Ethereum Foundation researcher
post-quantum cryptography software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
No Encryption Break Has Been Reported
The source describes no confirmed cryptographic break, no recovered private keys and no demonstrated attack on NIST’s post-quantum standards. It also does not identify the AI models used in the reported private testing, the protocols tested, or whether any results were independently replicated. The claim that AI systems are producing mathematics at a rate beyond human communities is the source author’s characterization, not a measured comparison established by the details provided.
It remains unclear how many of the 722 manuscripts will withstand expert review, whether any result will prove relevant to cryptography, and whether organizations have found undisclosed weaknesses. The reported correction to one withdrawn proof underlines why publication and initial claims should not be treated as validation.
As an affiliate, we earn on qualifying purchases.
Verification and Security Reviews Continue
The immediate next step is independent mathematical review of the published work and careful testing of any algorithmic claims with potential cryptographic relevance. The source provides no date for a completed review, no scheduled security milestone and no confirmed deadline for changes to standards.
Organizations responsible for finance, government and defence systems will need to distinguish between speculative risk and verified weaknesses as evidence develops. For blockchain users, Drake and Buterin’s contrasting advice points to a live debate over precautions, but the supplied material does not establish an official instruction to move funds. Future claims should be judged by reproducibility, technical detail and confirmation from cryptography specialists—not by the existence of AI-generated manuscripts alone.
As an affiliate, we earn on qualifying purchases.
Key Questions
Has AI broken encryption?
No break is reported in the source material. The manuscripts contain mathematical claims, and researchers are discussing whether AI could find algorithms relevant to cryptography.
What did OpenAI publish?
OpenAI published 722 mathematical manuscripts across 372 problem families on Oct. 6, generated by an unreleased internal model. The claims are subject to checking, and at least one reported proof was withdrawn after an error was found.
How is the AI concern different from the quantum threat?
A sufficiently capable quantum computer running Shor’s algorithm could threaten RSA and elliptic-curve systems. The AI concern is that a model might help discover a better algorithm that runs on ordinary computers; the source reports no such cryptographic result so far.
Are post-quantum standards affected?
The source raises questions about the assumptions behind lattice-based standards such as ML-DSA, but reports no attack or demonstrated weakness in them. NIST standardized ML-KEM, ML-DSA and hash-based SLH-DSA in August 2024.
Should cryptocurrency users move their funds?
Drake advocated planning for possible precautions, while Buterin said he did not recommend scrambling to move funds immediately. The source does not report a confirmed attack or an official requirement to move assets.
Source: ThorstenMeyerAI.com
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
