Unconditional Unclonable Encryption

Prabhanjan Ananth, Seyoon Ragavan, Amit Sahai

Ananth, Ragavan and Sahai give efficient, perfectly correct, one-time private-key encryption of one bit using a random nonidentity phase-free n-qubit Pauli and a random eigenstate. After key disclosure, both separated receivers recover the bit with probability at most 1/2+O(2^(−n/2)) against information-theoretic attacks.

Sources for this record

What was already known

Earlier unconditional schemes had inverse-polynomial security loss or inefficient encryption or decryption, according to the paper.

AI involvement AI-led

The authors credit independent Codex GPT-5.6 Sol Ultra conversations with the construction and proof. They supplied harness ideas, merged drafts, verified and substantially revised the argument.

The authors explicitly assign discovery of the central construction and proof to Codex across separate extended conversations, with human harness design and later revision.

Importance Notable

The construction combines unconditional security, efficiency and exponentially small cloning advantage for one-bit messages.

Limitations

The reported Lean proof covers Theorem 4.1 correctness and security, not efficiency or later arbitrary-message applications. Those extensions are separate human-led work.

Technical verification Formal (Lean 4 + human)

The paper reports careful author verification of the complete proof and says the public Lean repository proves all correctness and security claims in Theorem 4.1. The Lean scope excludes efficiency and downstream applications, which rest on the paper's separate human exposition and prior reductions. Neither check was rerun here.

Formal checking

Formalization of the operational scheme and complete Pauli security bound

Checker
Lean 4 on the authors' Pauli construction repository
Scope
Theorem 4.1 correctness and one-bit unclonable-indistinguishability security, excluding efficiency
Basis
Reported by the linked checker
Evidence source

Formal checking

Authors report careful verification and substantial cleanup of model-generated drafts

Checker
Prabhanjan Ananth, Seyoon Ragavan and Amit Sahai
Scope
The one-bit Pauli construction and proof of its security bound
Basis
Reported by the linked checker
Evidence source

Checking a formal statement does not establish that it matches the original problem.

Human understanding Full

The authors explain why expanding the key set to all nonidentity Paulis enables the negligible advantage and how operator bounds and disagreement terms complete the security proof.

Correction and publication history

2026-10-18 · publication

The construction combines unconditional security, efficiency and exponentially small cloning advantage for one-bit messages.

Read this published revision