Data provenance tracking (AI, IoT, etc.)

Can you prove where content or data came from, and that no one altered it? We build provenance systems that are hard to fake and easy to check, even when parts of the system are untrusted. Track A (AI-generated content): implement watermarking schemes whose keys are split among several parties, make every watermarking action publicly auditable, and use trusted execution environments to attest which model produced an output. Track B (sensor/IoT data): emulate a network of edge devices, attack its signed provenance records (replay, splicing, rollback, key misuse), and design verification at fog gateways that catches these attacks.

Name of research group, project, or lab
ND-DSP
Why join this research group or lab?

Why it matters. AI can now produce text, images and video that are hard to tell from human work. Military and infrastructure decisions also rely more and more on data from sensors placed where attackers can reach them. Both raise the same question: where did this come from, and has it been changed? Today’s answers usually require blind trust. An AI watermark is typically only as trustworthy as the single company holding its secret key, and anyone who steals that key can forge or strip watermarks. A signed sensor record helps only if replayed, spliced or rolled-back records get caught. We’re building provenance that holds up without blind trust.

 

Why our lab. The Data Security and Privacy Lab (DSP-Lab) works where security meets AI: homomorphic encryption, trusted hardware (Intel SGX, confidential VMs), blockchain-based data provenance and accountable AI. Undergraduates here are real team members:

  • You’ll get a structured onboarding package with a reading plan, hands-on exercises and self-check questions.
  • You’ll attend group meetings with Ph.D. students and work directly with me.
  • You’ll own a piece of a real system. E.g., our current undergraduates are building parts of a decentralized file-sharing platform and an encrypted large language model.
Logistics Information:
Project categories
Computer Science & Engineering
Student ranks applicable
Sophomore
Junior
Senior
Graduate Student
Student qualifications
  • Undergraduates in CSE, EE or a related major who program comfortably in Python and at least one of C/C++, Rust or Go, and who are at home on the Linux command line.
  • Curiosity about security, AI or systems, and steady weekly time for the project. No prior research or security experience is required; we’ll teach you.
  • Helpful but not required: a course in security, cryptography, operating systems or networking. For Track A, experience with PyTorch, Hugging Face or LLMs. For Track B, experience with Docker/VMs, networking or IoT/embedded systems.
  • Eligibility: Track B supports Air Force work that will be export-controlled, so it is open only to U.S. citizens and permanent residents. Track A has no citizenship requirement.
Hours per week
1 credit / 3-6 hours
2 credits / 6-12 hours
3 credits / 12+ hours
Compensation
Research for Credit
Number of openings
2
Techniques learned
  • Track A:
    • How LLM watermarks are embedded and detected, and how attackers remove or forge them.
    • Running and modifying open LLMs (PyTorch, Hugging Face).
    • Cryptographic building blocks: secret sharing, threshold keys, digital signatures, hash chains and Merkle-tree audit logs.
    • Trusted execution environments (Intel SGX, confidential VMs) and remote attestation.
  • Track B:
    • Threat modeling and implementing attacks: replay, splicing, rollback and key misuse.
    • TPM 2.0, software TPM emulation and tamper-evident signed records.
    • Emulating many networked devices with containers and VMs.
    • Measuring security: detection rate, false alarms, latency and overhead.
  • Both tracks: reading and presenting research papers, designing reproducible experiments, working as a team in Git, and technical writing. Strong work can lead to co-authorship on a paper.
Contact Information:
Mentor
tjung@nd.edu
Associate Professor
Name of project director or principal investigator
Taeho Jung
Email address of project director or principal investigator
tjung@nd.edu
2 sp. | 0 appl.
Hours per week
1 credit / 3-6 hours (+2)
1 credit / 3-6 hours2 credits / 6-12 hours3 credits / 12+ hours
Project categories
Computer Science & Engineering