MPC in the Wild

Workshop: MPC Security in Practice	Internships
Aarhus University	Aarhus University
18 May 2026	May-August 2026
co-located with TPMPC 2026

About

Secure multi-party computation (MPC) has been studied theoretically for decades now, and we are starting to see deployment. While MPC security guarantees have been well-established in theoretical models, the security of actual (open-source) implementations is currently understudied. We believe there is a disconnect between the theory of MPC, where protocols are designed and analyzed on paper, and the reality of MPC implementations.

MPC in the Wild is a research project that aims to study the real-world implementation security of MPC protocols, in particular, the gap between assumptions on the theory side and implementation choices made in practice.

Workshop: MPC Security in Practice

MPC Security in Practice is a 1-day workshop and will take place on 18 May 2026 at Aarhus University, Denmark as part of TPMPC 2026. The workshop will bring together speakers from MPC practice to discuss challenges for the security of MPC, as it is deployed in practice and the impact of these challenges on MPC theory. The program will consist of a mix of contributed and invited talks.

Confirmed speakers

• Martin Albrecht
• Sofía Celi
• Anders Dalskov
• Thomas Schneider

Program

tba

Contributed talks

We will solicit contributed talks on topics related to the project. Submissions will be handled as part of the TPMPC submission process, please refer to the TPMPC website for more details. The submission deadline is 2 March 2026 (AoE).

Registration

If you wish to join us for the MPC Security in Practice workshop, please register through the TPMPC website.

Visa and travel

Please refer to the TPMPC website for information regarding visas and traveling to Aarhus.

Internships

We are looking for multiple interns to work with us on the project during Summer 2026. Interns are expected to be current or recent PhD students with a relevant background in at least one of the following research areas:

• Modeling of MPC (security) in general, or of real-world aspects of other types of cryptographic protocols
• Attacks on implementations of advanced cryptographic protocols such as MPC, ZK, or related protocols
• Implementations of MPC protocols and related protocols

In rare cases, we may also consider master or bachelor students with relevant expertise.

Timeline

Interns are expected to participate in the full program. If you are interested but feel you cannot commit to the full duration (e.g. for medical reasons, care obligations, …) please mentions this in your application and we will see if we can find a solution.

Location

The internship will take place in the Cryptography and Security group at Aarhus University, Denmark.

How to apply

Apply for an internship

What we offer

Interns will be working on pre-agreed research projects with Peter Scholl and Sabine Oechsner, and possibly other members of the Cryptography and Security group. We will cover travel costs to and from Aarhus as well as a stipend to cover living expenses in Denmark. If needed, we can also offer support with visas and finding accommodation.

Research

Below, you can find some research results that led to or came out of this project:

• Rushing at SPDZ: On the Practical Security of Malicious MPC Implementations. Alexander Kyster, Frederik Huss Nielsen, Sabine Oechsner, and Peter Scholl. IEEE S&P 2025
• Covert Attacks on Machine Learning Training in Passively Secure MPC. Matthew Jagielski, Rahul Rachuri, Daniel Escudero and Peter Scholl
• Who Verifies the Verifiers? Lessons Learned From Formally Verified Line-Point Zero-Knowledge. Sabine Oechsner, Vitor Pereira, and Peter Scholl. Communications in Cryptology, 2025

Team

MPC in the Wild is run by

• Sabine Oechsner (VU Amsterdam) and
• Peter Scholl (Aarhus University).

Research interns (summer 2026):

• Nan Cheng (University of St. Gallen)
• Bhavish Raj Gopal (IISc Bangalore)
• Miquel Guiot (Universitat Rovira i Virgili)
• Kaiwen (Kevin) He (MIT)

Support

The project is supported by a Cryptanalysis Grant from Zama.