Final remark:

Since the threat comes from violating a Bell-type inequality (specifically, Mermin inequality), it has a purely quantum nature:

(a) It makes no computational assumption
(b) It doesn't come from additional communication channel, as entanglement can't transmit information.

5/5

While the scenario is specific, it reveals a threat from quantum entanglement to access control, showing existing models are insufficient.

To protect against the threat, we design new quantum access control models, and analyze their security, flexibility and efficiency.

4/5

We show the answer is likely *no*.

Intuition:

Access control governs how users are allowed to access resources. If a system's security relies on a Bell-type inequality, and the access control mechanism allows users to test it, then introducing quantum resources can cause a security breach.

3/5

Motivation:

You trust a classical computer system, as its access control mechanism is *proven* to protect your private information. One day, the system upgrades by integrating quantum computing services. Should you still trust this system?

2/5

(b) "Quantum Register Machine: Efficient Implementation of Quantum Recursive Programs" (arxiv.org/abs/2408.10054)

(4/4)

It's based on the following two joint works with Mingsheng Ying. I'll also present paper (b) at PLDI 2025 in June.

(a) "Verification of Recursively Defined Quantum Circuits" (arxiv.org/abs/2404.05934)

(3/4)

Content:
- Quantum & PL background
- Syntax & Semantics of RQC++, a quantum recursive programming language
- Various examples
- A theoretical framework for efficiently implementing quantum recursive programs

(2/4)

Thanks a lot Clément for this post!

Can I be added please? Thanks!