Aydin AbadiProtocol design · Formal security · Cryptanalysis
Menu
Cryptography and AI Security Lab

Cryptography and AI Security Lab

Collaborate with the lab

CASL collaborates with organisations and researchers on defined problems in cryptographic protocol design and analysis, including sensitive-data computation, private access, verifiable computation and secure digital exchange.

Send an enquiry

Research fit

Problems within the lab’s scope

  • Computing or learning across sensitive datasets without pooling raw records.
  • Accessing remote data, databases or machine-learning models privately.
  • Verifying delegated or time-delayed computation under explicit trust assumptions.
  • Designing payment or service-exchange protocols with verifiable delivery and dispute handling.
  • Analysing proposed cryptographic protocols and security arguments for exploitable assumptions or attacks.

Collaboration routes

How we can work together

The appropriate route depends on the research question, the maturity of the system and the intended output.

Suitable projects can be arranged through Newcastle University as collaborative research, commercial consultancy or a Knowledge Transfer Partnership. Short technical workshops may also support a defined research or knowledge-exchange objective.

School of Computing partnership routes Newcastle collaborative research

Relevant evidence

Research behind the collaboration areas

Selected work related to private data collaboration, private access and verifiable outsourcing.

Financial data collaboration

STARLIT

Privacy-preserving federated learning for financial fraud detection without pooling raw transaction data. STARLIT was a joint first-place winner in the UK-US PETs Prize Challenge.

Read the STARLIT paper View the prize result

Private access · Constrained clients

Oblivis

Delegated oblivious transfer under a non-collusion assumption. Supersonic OT achieves at least a 92× speedup and completes an end-to-end transfer in 1.36 ms on a Raspberry Pi 4.

Read the paper

Verifiable delegated computation

Scalable Time-Lock Puzzle

A delegated time-lock framework with verifiable computation for clients and servers with substantially different computational capabilities.

View the publication

Initial email

Information to include

  1. Describe the use case or system.
  2. Name the sensitive assets or security objective.
  3. Identify the relevant threat actors or trust limits.
  4. State the current stage and desired timescale.
  5. Say what outcome would be useful.

Possible outputs

What a project may produce

  • A defined threat model and precise security objectives.
  • A protocol specification and formal security analysis.
  • A proof-of-concept research implementation.
  • Benchmarking or a cryptographic feasibility study.
  • A joint funding application or academic publication, where appropriate.

The precise output depends on the question and the agreed University route. Research prototypes are intended for feasibility and evaluation rather than production deployment.

Email aydin.abadi@ncl.ac.uk

Contact

Research and technical enquiries

Send a concise description of the problem, system and intended output. We can then assess whether it aligns with the lab's expertise.

Email AydinExplore the research