Post-Quantum Transition Henry Quentir Post-Quantum Transition Henry Quentir

How Many Servers Offer Post-Quantum Key Exchange? Two arXiv Scans of 6 November 2026 Count 45.2% of SSH Hosts and 9.1% of Scanned HTTPS Servers, Where F5 Labs Counts 54% of Websites

What two arXiv papers of 6 November 2026 found on SSH and TLS servers

Md Ariful Haque, Sander Vinberg, Ben Edwards and João Cruz posted two measurement studies to arXiv on 6 November 2026, both built on Bitsight's GROMA internet scanning platform. The SSH paper counts 688,814 of 1,523,998 hosts on port 22, or 45.2%, offering a post-quantum key exchange in a scan of 16 February 2026. The TLS paper finds the hybrid X25519MLKEM768 group selected in its own test connections by 9.11% of the HTTPS server addresses it reached and 7.39% of LDAPS servers, and by fewer than 1% of servers for SMTPS, IMAP, POP3, TELNETS and RDP.

Why the percentages disagree with F5 Labs and Forescout

F5 Labs' 2026 report of 22 September counts 54% of the top million websites supporting post-quantum key exchange, and 22% once Cloudflare-hosted domains are removed. Forescout counted 11.8% of about 160 million SSH servers in June. Each figure uses a different population and a different test: an offered algorithm or a negotiated one, a content-delivery edge or an individual server address. Most of the SSH result also rests on sntrup761, a hybrid NIST did not standardize; a hybrid incorporating ML-KEM appears on 5.79% of hosts. The studies are not directly comparable, and definitions explain only part of the spread, which is why a shared definition of PQC adoption matters for supervisors, auditors and buyers.

Read More
Post-Quantum Transition Henry Quentir Post-Quantum Transition Henry Quentir

NIST's Final IR 8547 of 27 October 2026 Deprecates P-256, RSA-3072 and Ed25519 After 2030, Where the 2024 Draft Left 128-Bit Keys Untouched Until 2035

What NIST changed in the final IR 8547 on 27 October 2026

NIST published the final version of NIST IR 8547, Transition to Post-Quantum Cryptography Standards, on 27 October 2026, replacing the initial public draft of 12 November 2024. The central change sits in the transition tables. The draft deprecated only quantum-vulnerable algorithms at 112 bits of security after 2030 and let stronger keys run until they were disallowed after 2035. The final deprecates every quantum-vulnerable public-key algorithm at 112 bits or more after 2030, so ECDSA on P-256, RSA-3072 and Ed25519 now enter the deprecated category at the same point as RSA-2048. Disallowance after 2035 stays where it was.

How the final ties NIST standards to EO 14412 and OMB M-26-15

The report now cites Executive Order 14412 of 22 June 2026 and OMB Memorandum M-26-15 of 24 June 2026. Under them, federal high value assets and high impact systems must use post-quantum key establishment by 31 December 2030 and post-quantum signatures by 31 December 2031, and agencies must support TLS 1.3 by 2 January 2030. NIST says the 2030 to 2035 window exists for lower-risk and legacy systems and does not relax those deadlines. The final also names ML-KEM-768 as the default parameter set, confirms HQC as the second key-establishment scheme and states that an approved composite, or hybrid, key-establishment technique can be FIPS 140 validated when at least one of its inputs comes from an approved scheme.

Read More
Post-Quantum Transition Henry Quentir Post-Quantum Transition Henry Quentir

ML-KEM Has Moved Into the Hardware Test Lab

The standard is now a device

Post-quantum cryptography has crossed an awkward threshold. ML-KEM is no longer only a standards document, a migration milestone or a line item in a crypto-agility plan. Once it lands in hardware, firmware and embedded libraries, its security also depends on power traces, electromagnetic leakage and the exact sequence of operations during decapsulation. A new 30 June 2026 arXiv paper on Fujisaki-Okamoto verification in ML-KEM makes that point concrete: the verification step can become a visible leakage surface during physical side-channel analysis.

Why procurement changes

The useful commercial lesson is narrow and important. Buyers should not treat ML-KEM implementation security as a checkbox created by adopting a NIST algorithm name. They need to know whether their chips, HSMs, gateways, telecom equipment, IoT modules and cloud cryptographic services have been tested against the way the algorithm runs in the real device. That moves post-quantum transition work closer to product assurance, certification, warranty drafting and supplier disclosure.

Quentir’s reading

This does not weaken the case for migration. It sharpens it. The next mature post-quantum program will connect algorithm selection with side-channel assurance, validated components, patch rights, test reports and contractual responsibility when a “quantum-safe” implementation leaks through the hardware layer. That is where policy deadlines become operational.

Read More