Saltgrain and NTT Research Develop Large-Universe Post-Quantum Encryption

Saltgrain, NTT Research, and Carnegie Mellon have released a peer‑reviewed large‑universe post‑quantum attribute‑based encryption scheme with millisecond decryption on a laptop.

Saltgrain and NTT Research Develop Large-Universe Post-Quantum Encryption

Large‑Universe Post‑Quantum Attribute‑Based Encryption: From Theory to Practice

Saltgrain, the data‑centric security platform incubated at NTT Research’s Scale Academy, together with researchers from NTT Research’s Cryptography and Information Security (CIS) Lab and Carnegie Mellon University, have announced the first practical large‑universe post‑quantum attribute‑based encryption (ABE) schemes. The work, presented at ACM CCS 2026 and described in a peer‑reviewed paper (ia.cr/2026/1895), extends lattice‑based cryptography to support fine‑grained, role‑based access control that can scale to arbitrary attribute sets while resisting attacks from future quantum computers.

---

What Makes This ABE Scheme “Large‑Universe”?

Traditional ABE systems require a fixed list of attributes to be defined at setup, limiting flexibility as organizations evolve. The new schemes introduce a large‑universe model: any attribute label can be added later without re‑issuing keys or redefining the system. A multi‑authority variant also allows separate departments or partner organisations to issue credentials independently, simplifying cross‑organisational data sharing.

Both key‑policy (KP‑ABE) and ciphertext‑policy (CP‑ABE) constructions are built on the Learning With Errors (LWE) problem and its ring variant— the same hardness foundation behind NIST’s post‑quantum standardisation efforts. While the schemes themselves are not yet NIST‑certified, they inherit the same 128‑bit security level claimed for the underlying lattice primitives.

---

Benchmarks: From Theory to an Ordinary Laptop

The research team measured performance on an Apple M4 Pro MacBook Pro, encrypting a 256‑bit AES key under AND policies of one to four attributes. Reported timings are:

  • Decryption: 5.6 ms (KP‑ABE, 1 attribute) to 22.5 ms (KP‑ABE, 4 attributes); 7.5 ms to 33.5 ms for CP‑ABE.
  • Encryption: 195 ms to 545 ms for the KP scheme.
  • Ciphertext size: Ring‑LWE versions pack 2 048 message bits per ciphertext, yielding roughly a 2 048‑fold reduction in per‑bit overhead compared with one‑bit constructions.

These numbers place the scheme in the “practical” range for enterprise use‑cases where a file is unlocked in a few hundredths of a second, a stark contrast to earlier post‑quantum ABE prototypes that required seconds to minutes.

---

Integration Path with Saltgrain

Saltgrain plans to embed the new ABE primitives into its platform, allowing organisations to attach quantum‑resistant, attribute‑based access policies directly to data objects. The approach complements existing security controls— it does not replace firewalls, IAM, or DLP solutions but adds a cryptographic layer that persists even if data is exfiltrated.

No commercial availability date was announced, but the press release emphasises that the technology is already undergoing integration testing. For CISOs, the promise is clear: protect long‑lived, high‑value data today against adversaries who might harvest it now and decrypt it later with a quantum computer.

---

Outlook

The announcement arrives at a pivotal moment: NIST’s post‑quantum standards were finalised in 2024, and U.S. executive orders set migration deadlines for 2030‑31. By delivering a usable, large‑universe ABE construction, Saltgrain and its partners provide a concrete building block for organisations seeking crypto‑agility without sacrificing fine‑grained access control.

Future work will likely focus on expanding attribute complexity (e.g., OR policies, hierarchical attributes) and reducing ciphertext size further, as well as obtaining independent security audits and real‑world pilot deployments.

---

For more details, see the paper (ia.cr/2026/1895), the NTT Research blog post, and the accompanying video.

Hype check

Real

The headline accurately reflects a peer‑reviewed research result with concrete benchmark data; the paper and conference presentation provide sufficient evidence.

  • Peer-reviewed paper, or press release only?Paper — Paper cited (ia.cr/2026/1895) presented at ACM CCS 2026
  • Error-corrected (logical) qubits, or physical qubits only?Logical / not a qubit claim — Not a qubit claim; question not applicable
  • A new milestone, or a rebrand / restatement?New — First practical large‑universe post‑quantum ABE reported
  • Independent confirmation?Not yet — No external validation mentioned beyond authors
  • Does the headline claim match what the source shows?Matches — Headline matches the described development and results

An editorial assessment of the claim as reported, not a judgment of the company or researchers. Checked against: ntt-research.com, youtu.be. How we rate hype →

Source

Read original source

Why we picked this

Accepted from curated QuantumUnzipped source (relevance scoring disabled).

← Back to all articles