Tratopedia
Side N繁中
Settings

Text size

Language

Theme

High contrast

Version

v1.230.2

The release this page was built from. It is what the service worker caches under.

Cryptography · Language Models · AI Capability · Source Verification · Research BriefAnthropic's announcement of 28 Jul 2026 · record to 28 Aug 2026

Executive summary · one-page brief

Read the Papers, Not Just the Post

Anthropic says its unreleased Claude Mythos Preview model produced two genuine cryptanalytic results: an improved key-recovery attack on the HAWK post-quantum signature scheme, and a faster attack on 7-round AES-128. Both are published as full papers, and both are real. But each paper also states a second, less flattering number for the same result that Anthropic's own blog post does not carry — a gap an independent reviewer of both documents has already pointed out in public.

The two results, and a benchmark Neither threatens a deployed system

  • 200–800×AES-128 (7 rounds) speed-up Anthropic's post quotes — at a still-impractical data cost
  • 2.7bits — the same paper's own balanced-complexity gain over the 2013 baseline
  • 2^108 / 2^182HAWK-512 / HAWK-1024 key-recovery cost in gates, down from 2^150 / 2^288 under the specification's own model
  • 85.7%CryptanalysisBench Tier-1 score for the model behind both attacks — highest of five models tested

Standing Sorted by how well each claim is evidenced

StandingWhat happenedWhen
ConfirmedAnthropic publishes a blog post and two full technical papers describing results from an unreleased model, Claude Mythos Preview: an improved HAWK key-recovery attack, and a faster 7-round AES-128 attack.28 Jul 2026
ConfirmedThe HAWK paper (Straznickas & Weis) reduces key recovery to poly(n) calls to an SVP oracle in dimension n/2+1 via a previously unused lattice automorphism, cutting HAWK-512 / HAWK-1024's gate-count cost from 2^150 / 2^288 to at most 2^108 / 2^182 under the specification's own model, and recovers two HAWK-256 test keys end-to-end.25 Jul 2026
ConfirmedThe AES paper (Nasr & Carlini) removes one of nine guessed key bytes from a 2013 7-round AES-128 attack, bringing runtime at the same 2^105-plaintext data budget to between 2^89.3 and 2^91.4 — though the same paper's own balanced-complexity metric puts the actual gain at only 2.7 bits.28 Jul 2026
ConfirmedCryptanalysisBench, a companion benchmark built with academics at ETH Zurich, Tel Aviv University, the University of Haifa and TU Berlin, scores Mythos 5 at 85.7% on 49 known-broken schemes and 8.9% on 142 full-strength schemes with no known break — the highest of five models tested on both.20/29 Jul 2026
Confirmed but not verifiable hereNIST cryptographer Daniel Apon replies on the NIST pqc-forum mailing list within hours, saying the HAWK result checks out independently for him — as reported by one independent reviewer of the announcement.28 Jul 2026

From a 2013 baseline to this month's record Dated events only

  1. 2013Derbez, Fouque & Jean publish the previous-best 7-round AES-128 attack: 2^105 chosen plaintexts, 2^99 time.
  2. 2022HAWK is first proposed and later enters NIST's post-quantum digital-signature standardization process.
  3. May 2026NIST advances HAWK — one of nine candidates, and the only lattice-based one — to the third round of that process.
  4. Jun 2026Anthropic notifies HAWK's authors of the Mythos Preview result ahead of publication.
  5. 20 Jul 2026CryptanalysisBench is first posted to arXiv.
  6. 28 Jul 2026Anthropic publishes its blog post and both technical papers; HAWK co-author Stephen Weis posts the announcement to the NIST pqc-forum mailing list the same day, and NIST cryptographer Daniel Apon reportedly replies within hours confirming the HAWK result independently.
  7. 29 Jul 2026Anthropic edits its blog post (“updated an academic affiliation”); a revised CryptanalysisBench (v2) is posted; an independent analysis of both papers is published on PostQuantum.com.
  8. 28 Aug 2026As of this date, the public record does not show whether NIST or HAWK's submitters have changed the scheme's parameters or its standing in the standardization process.

Anthropic's own post names neither a month for HAWK's advance to the third round nor who replied on the NIST mailing list; both details come from independent reporting on the announcement — The Hacker News and Marin Ivezic's review, respectively.

  • 9candidates NIST advanced to the third round; HAWK is the only lattice-based one
  • 2/2HAWK-256 reference-generated public keys the released code recovered end-to-end

What Anthropic's post says The headline framing

Anthropic's post frames both results in their most striking terms. On HAWK, it says the attack “effectively cutting its key strength in half,” and states the expected work factor for the smaller HAWK-256 parameter set as falling from 2^64 to 2^38. On AES, it says Mythos improved the speed of the previous best attacks by “200-800×.” Both numbers are real and both trace to the underlying papers — the 2^38 and the 200-800× figures each match a number the papers themselves compute.

What the papers' own tables say A second, less flattering number

Neither paper stops at its headline number. The HAWK paper's own Table 1 — using the gate-count model the HAWK specification itself relies on — gives HAWK-256's spec-side figure as 2^74, not 2^64; the 2^62 figure the paper gives under a different model (Core-SVP) is the closest match, and the 2^64 pairing in Anthropic's post does not appear in the table at all. The AES paper is explicit that its 200-800× figure holds only the data requirement fixed at 2^105 — itself still impractical — and that under the standard balanced metric max(D,T,M), which weighs data, time and memory together, the true gain over the 2013 baseline is 2.7 bits (or 2.1 bits, without one of the paper's own fingerprint assumptions). The paper states this about its own result: “the overall complexity of our attack has not improved” under the fixed-data framing alone.

Headline against table, side by side Same result, two numbers

ClaimAnthropic's postThe paper's own table
HAWK-256 work factor2^64 → 2^382^74 → 2^52 (gate model) / 2^62 → 2^38 (Core-SVP)
AES-128 (7 rounds) speed-up200–800×2.7 bits (2.1 without one fingerprint assumption), balanced metric max(D,T,M)
HAWK-512 / HAWK-1024 key strength“cut in half”2^150 → 2^108 (42 bits) / 2^288 → 2^182 (106 bits)

None of this makes Anthropic's post false. Marin Ivezic, an independent security consultant who reviewed both papers, reaches the same reading: “the blog is a summary; the paper is the citable source, and the secondary coverage has been copying the blog.” The two attacks are real, published, and reviewable; the number a reader remembers depends on which document they read.

Three outside readings Security press, an independent analyst, and the paper's own acknowledgements

  • The Hacker News

    A Candidate Doing Its Job

    • HAWK is the only lattice-based scheme among the nine NIST advanced to the third round in May 2026.
    • Anthropic's released code recovers a 592-byte decoded key, not the original 96-byte secret seed, and works only against HAWK-256.
    • A 2025 paper by van Gent & Pulles had already shown a hypothetical automorphism would open this attack path, but its own authors said it did not then threaten HAWK; Mythos found the missing automorphism.
  • PostQuantum.com

    The Process Worked

    • Marin Ivezic notes Anthropic notified HAWK's team in June, posted to the NIST forum the same day as publication, and got an independent same-day check from a NIST cryptographer — disclosure norms followed, as with SIKE's 2022 break of a different candidate.
    • His guidance for security teams: leave standardized-algorithm roadmaps (ML-KEM, ML-DSA, SLH-DSA, Falcon) alone — none is affected — and instead inventory any proprietary or vendor-specific cryptography, which is the population these results actually bear on.
    • He calls the verification bottleneck — not either specific break — the thing worth watching, and cites cryptographer Markku-Juhani Saarinen's forum proposal that AI-assisted cryptanalysis should always ship machine-checkable proofs or scaled-down demonstrations.
  • The AES paper's acknowledgements

    Six Names, Comments Only

    • Nasr & Carlini thank Orr Dunkelman, Patrick Derbez, Jérémy Jean, Eyal Ronen, Nathan Keller and Adi Shamir — and say only that they gave “comments on an early draft.”
    • Three of the six — Shamir, Dunkelman and Keller — are named co-authors of the 2010 and 2020 papers this attack directly extends; two more, Derbez and Jean, co-authored the 2013 baseline it improves on.
    • The acknowledgement names no endorsement, no review, and no verification — only comments on a draft, and nothing further is claimed on their behalf.

The verification bottleneck The papers' own authors raise it

Nasr & Carlini describe the core AES idea as taking “tens of hours” of model time against “hundreds of hours” of their own validation, and call the resulting arrangement — humans mainly checking a model's output rather than producing it — something that “feels uncomfortable.” Ivezic treats that ratio, not either specific break, as “the thing I would put on a watch list,” describing a field that can now “discover” faster than it can review.

What stays true regardless The boundaries both sides agree on

HAWK is a candidate, not a deployed standard; the construction does not transfer to Falcon and does not touch ML-KEM, ML-DSA or SLH-DSA. Full 10-round AES-128 is untouched — the attack targets 7 rounds and still needs an impractical 2^105 chosen plaintexts. And CryptanalysisBench's own authors caution that because every Tier-1 scheme already has a published break, a model's “genuine”-looking win “does not by itself establish independent rediscovery” over recall of a known attack — roughly 78% of wins across all five tested models were judged genuine by the paper's own trace analysis, a proportion nearly identical across models rather than a Mythos-specific figure.

So what Real advances, read at the right resolution

What's solid

Two published, reviewable cryptanalytic advances against a not-yet-deployed post-quantum candidate and a reduced-round cipher already impractical at full strength, plus a benchmark on which the same model leads four rivals. No production system needs to change.

What's overstated

The two figures a reader is likeliest to remember — HAWK-256's “2^64 to 2^38” and AES's “200-800×” — are each the most favourable framing available, and each paper states a second, smaller number for the same result that its own announcement does not carry.

What's still open

Whether a model that discovers this fast can be verified this fast is a question the papers' own authors raise about themselves and do not answer. Whether NIST or HAWK's submitters respond by changing the scheme is, as of 28 Aug 2026, not yet shown in the public record either way.

  • Watch whether NIST or HAWK's submitters respond — a parameter change, a withdrawal, or silence through the next standardization update would each say something different.
  • Watch whether CryptanalysisBench's Tier-1 lead for Mythos-class models holds at longer time budgets — the paper's own 16-hour run already found most of that gain came from finishing attacks already found, not from new capability.
  • Watch whether machine-checkable proofs or scaled-down demonstrations — Saarinen's proposal — become standard practice for AI-assisted cryptanalytic claims generally.
  • Watch for any of the six cryptographers thanked in the AES paper's acknowledgements publishing their own assessment of the result, rather than the unquoted “comments” the paper itself reports.

Quick notes

  1. Cryptography · Research brief

    Read the Papers, Not Just the Post

    Two real, published cryptanalytic advances — and in each paper a second, less flattering number that Anthropic's blog post does not carry.

    Cryptographic weakness discovery1 / 17

  2. What Happened

    The two results, and a benchmark

    Neither threatens a deployed system

    • 200–800×AES-128 (7 rounds) speed-up Anthropic's post quotes — at a still-impractical data cost
    • 2.7bits — the same paper's own balanced-complexity gain over the 2013 baseline
    • 2^108 / 2^182HAWK-512 / HAWK-1024 key-recovery cost in gates, down from 2^150 / 2^288 under the specification's own model
    • 85.7%CryptanalysisBench Tier-1 score for the model behind both attacks — highest of five models tested

    Cryptographic weakness discovery2 / 17

  3. What happened · the HAWK paper

    An improved key-recovery attack on HAWK

    Straznickas & Weis · 25 Jul 2026

    The method

    Key recovery reduces to poly(n) calls to an SVP oracle in dimension n/2+1, via a previously unused lattice automorphism.

    The cost

    HAWK-512 / HAWK-1024 fall from 2^150 / 2^288 to at most 2^108 / 2^182 gates, under the specification's own model.

    The demonstration

    The released code recovered two HAWK-256 test keys end-to-end — a 592-byte decoded key, not the original 96-byte secret seed, and only against HAWK-256.

    Cryptographic weakness discovery3 / 17

  4. What happened · the AES paper

    One of nine guessed key bytes removed

    Nasr & Carlini · 28 Jul 2026 · building on Derbez, Fouque & Jean's 2013 attack on 7-round AES-128

    • 2^99Time of the 2013 baseline, at 2^105 chosen plaintexts
    • 2^89.3–2^91.4Runtime now, at the same 2^105 data budget — the basis of the 200–800× figure
    • 2.7bits — the gain under the balanced metric max(D,T,M) (2.1 without one fingerprint assumption)

    Cryptographic weakness discovery4 / 17

  5. What happened · CryptanalysisBench

    A benchmark on which the same model leads four rivals

    Built with academics at ETH Zurich, Tel Aviv University, the University of Haifa and TU Berlin

    • 85.7%Mythos 5 on 49 known-broken schemes
    • 8.9%Mythos 5 on 142 full-strength schemes with no known break
    • The highest of the five models tested, on both.
    • Every Tier-1 scheme already has a published break, so a genuine-looking win “does not by itself establish independent rediscovery”, its own authors caution.

    Cryptographic weakness discovery5 / 17

  6. Timeline

    From a 2013 baseline to the announcement

    Dated events only

    1. 2013Derbez, Fouque & Jean publish the previous-best 7-round AES-128 attack: 2^105 chosen plaintexts, 2^99 time.
    2. 2022HAWK is first proposed and later enters NIST's post-quantum digital-signature standardization process.
    3. May 2026NIST advances HAWK — one of nine candidates, and the only lattice-based one — to the third round of that process.
    4. Jun 2026Anthropic notifies HAWK's authors of the Mythos Preview result ahead of publication.
    5. 20 Jul 2026CryptanalysisBench is first posted to arXiv.

    Cryptographic weakness discovery6 / 17

  7. Timeline

    Announcement day, and the month after

    1. 28 Jul 2026Anthropic publishes its blog post and both technical papers; HAWK co-author Stephen Weis posts the announcement to the NIST pqc-forum mailing list the same day, and NIST cryptographer Daniel Apon reportedly replies within hours confirming the HAWK result independently.
    2. 29 Jul 2026Anthropic edits its blog post (“updated an academic affiliation”); a revised CryptanalysisBench (v2) is posted; an independent analysis of both papers is published on PostQuantum.com.
    3. 28 Aug 2026As of this date, the public record does not show whether NIST or HAWK's submitters have changed the scheme's parameters or its standing in the standardization process.

    Cryptographic weakness discovery7 / 17

  8. The Argument

    What Anthropic's post says

    The headline framing

    HAWK

    “effectively cutting its key strength in half”; HAWK-256's expected work factor falls from 2^64 to 2^38.

    AES

    Mythos improved the speed of the previous best attacks by “200-800×.”

    Both are real

    The 2^38 and the 200-800× figures each match a number the papers themselves compute.

    Cryptographic weakness discovery8 / 17

  9. The Argument

    What the papers' own tables say

    A second, less flattering number

    HAWK-256, Table 1

    Under the gate-count model the HAWK specification relies on, the spec-side figure is 2^74, not 2^64; under Core-SVP it is 2^62. The pairing “2^64” does not appear in the table at all.

    AES, balanced metric

    200-800× holds only with data fixed at 2^105, itself still impractical. Under the balanced metric max(D,T,M), the gain over the 2013 baseline is 2.7 bits (2.1 without one fingerprint assumption).

    In the AES paper's own words

    “the overall complexity of our attack has not improved” — under the fixed-data framing alone.

    Cryptographic weakness discovery9 / 17

  10. The Argument

    Headline against table, side by side

    Same result, two numbers

    ClaimAnthropic's postThe paper's own table
    HAWK-256 work factor2^64 → 2^382^74 → 2^52 (gate model) / 2^62 → 2^38 (Core-SVP)
    AES-128 (7 rounds) speed-up200–800×2.7 bits (2.1 without one fingerprint assumption), balanced metric max(D,T,M)
    HAWK-512 / HAWK-1024 key strength“cut in half”2^150 → 2^108 (42 bits) / 2^288 → 2^182 (106 bits)

    Cryptographic weakness discovery10 / 17

  11. The Argument

    None of this makes Anthropic's post false

    Marin Ivezic, independent security consultant

    “the blog is a summary; the paper is the citable source, and the secondary coverage has been copying the blog.”

    Which document you read

    The two attacks are real, published, and reviewable; the number a reader remembers depends on which document they read.

    Cryptographic weakness discovery11 / 17

  12. What Others Add

    Three outside readings

    The Hacker News · A candidate doing its job

    HAWK is the only lattice-based scheme among the nine NIST advanced to the third round. A 2025 paper by van Gent & Pulles showed a hypothetical automorphism would open this path, but its authors said it did not then threaten HAWK; Mythos found the missing automorphism.

    PostQuantum.com · The process worked

    Marin Ivezic notes Anthropic notified HAWK's team in June, posted to the NIST forum on publication day, and got a same-day independent check from a NIST cryptographer — disclosure norms followed.

    The AES paper's acknowledgements · Six names, comments only

    Nasr & Carlini thank six cryptographers, including Adi Shamir, only for “comments on an early draft” — no endorsement, no review, no verification.

    Cryptographic weakness discovery12 / 17

  13. What Others Add · PostQuantum.com

    Marin Ivezic's guidance for security teams

    Leave alone

    Standardized-algorithm roadmaps — ML-KEM, ML-DSA, SLH-DSA, Falcon. None is affected.

    Inventory instead

    Any proprietary or vendor-specific cryptography — the population these results actually bear on.

    Cryptographic weakness discovery13 / 17

  14. What Others Add

    The verification bottleneck

    The papers' own authors raise it

    • Tens of hoursof model time for the core AES idea
    • Hundreds of hoursof the authors' own validation
    • Nasr & Carlini call the arrangement — humans mainly checking a model's output rather than producing it — something that “feels uncomfortable.”
    • Ivezic calls that ratio, not either specific break, “the thing I would put on a watch list”, and cites Markku-Juhani Saarinen's proposal that AI-assisted cryptanalysis should always ship machine-checkable proofs or scaled-down demonstrations.

    Cryptographic weakness discovery14 / 17

  15. What Others Add

    What stays true regardless

    The boundaries both sides agree on

    HAWK

    A candidate, not a deployed standard. The construction does not transfer to Falcon and does not touch ML-KEM, ML-DSA or SLH-DSA.

    AES

    Full 10-round AES-128 is untouched: the attack targets 7 rounds and still needs an impractical 2^105 chosen plaintexts.

    CryptanalysisBench

    Roughly 78% of wins across all five models were judged genuine by the paper's own trace analysis — nearly identical across models, not a Mythos-specific figure.

    Cryptographic weakness discovery15 / 17

  16. Conclusion

    So what

    Real advances, read at the right resolution

    What's solid

    Two published, reviewable cryptanalytic advances against a not-yet-deployed post-quantum candidate and a reduced-round cipher already impractical at full strength, plus a benchmark on which the same model leads four rivals. No production system needs to change.

    What's overstated

    The two figures a reader is likeliest to remember — HAWK-256's “2^64 to 2^38” and AES's “200-800×” — are each the most favourable framing available, and each paper states a second, smaller number for the same result that its own announcement does not carry.

    What's still open

    Whether a model that discovers this fast can be verified this fast is a question the papers' own authors raise about themselves and do not answer. Whether NIST or HAWK's submitters respond by changing the scheme is, as of 28 Aug 2026, not yet shown in the public record either way.

    Cryptographic weakness discovery16 / 17

  17. What to watch

    Four things to watch

    • Watch whether NIST or HAWK's submitters respond — a parameter change, a withdrawal, or silence through the next standardization update would each say something different.
    • Watch whether CryptanalysisBench's Tier-1 lead for Mythos-class models holds at longer time budgets — the paper's own 16-hour run already found most of that gain came from finishing attacks already found, not from new capability.
    • Watch whether machine-checkable proofs or scaled-down demonstrations — Saarinen's proposal — become standard practice for AI-assisted cryptanalytic claims generally.
    • Watch for any of the six cryptographers thanked in the AES paper's acknowledgements publishing their own assessment of the result, rather than the unquoted “comments” the paper itself reports.

    Cryptographic weakness discovery17 / 17

  • Anthropic, “Discovering cryptographic weaknesses with Claude”, 28 Jul 2026 (edited 29 Jul)
  • Straznickas & Weis, “HAWK-n Key Recovery Reduces to SVP in Dimension n/2+1”, Anthropic
  • Nasr & Carlini, “Cryptanalysis of 7-Round AES via the Algebraic Structure of its S-box”, Anthropic
  • Fluri, Shafran, Carlini, Jagielski, Nasr, Dunkelman, Ronen & Tramèr, “CryptanalysisBench: Can LLMs do Cryptanalysis?”, arXiv:2607.18538 (v2, 29 Jul 2026)
  • Khandelwal, The Hacker News, 28 Jul 2026
  • Ivezic, PostQuantum.com, 29 Jul 2026. All read 28 Aug 2026.

Reliable

Reliable, but with an interestWorth believing and reliableWeak on both, and used with careHonest, but thin

Worth believing

Each point is one source used, placed by two judgements on a six-step scale. Points carry no names: the chart shows the shape of the evidence, not a verdict on any publisher.

Versions

This document is rewritten when what it says has to change. Every version stays published at its own address.

  1. v0002 current
  2. v0001 superseded

The current version is also at latest/.