David Hofer

I’m a research engineer at Snyk, working on AI security, measuring and mitigating the risks of LLMs and AI agents.

Previously: MSc in Computer Science (focus Cyber Security) at ETH Zürich and EPFL.

news

Sep 21, 2026	New preprint: Beyond Predictable Paths: Redefining AI Security Incident Reporting for Agents
Aug 19, 2026	New blog post on the Snyk Labs research blog: How We Evaluate Model Risk
Jun 09, 2026	New preprint: Assessing Automated Prompt Injection Attacks in Agentic Environments
Feb 02, 2026	Starting a new position at Invariant Labs/Snyk, working on AI security
Dec 25, 2025	Graduated with a MSc in Computer Science (focus Cyber Security) from ETH Zurich & EPF Lausanne

latest posts

Aug 19, 2026	How We Evaluate Model Risk

selected publications & preprints

A Neural Model for Regular Grammar Induction

Peter Belcak, David Hofer, and Roger Wattenhofer

In 2022 21st IEEE International Conference on Machine Learning and Applications (ICMLA), Dec 2022

DOI arXiv Bib PDF

@inproceedings{10069747,
  author = {Belcak, Peter and Hofer, David and Wattenhofer, Roger},
  booktitle = {2022 21st IEEE International Conference on Machine Learning and Applications (ICMLA)},
  title = {A Neural Model for Regular Grammar Induction},
  year = {2022},
  month = dec,
  volume = {},
  number = {},
  pages = {401-406},
  keywords = {Training;Turing machines;Computational modeling;Machine learning;Programming;Natural language processing;Inference algorithms;neural networks;regular languages;grammar induction;program synthesis},
  doi = {10.1109/ICMLA55696.2022.00064},
}

Beyond Predictable Paths: Redefining AI Security Incident Reporting for Agents

Anastasia Pustozerova, Eugene Bagdasarian, Luca Beurer-Kellner, and 22 more authors

Sep 2026

arXiv Bib PDF

@misc{pustozerova2026beyond,
  title = {Beyond Predictable Paths: Redefining AI Security Incident Reporting for Agents},
  author = {Pustozerova, Anastasia and Bagdasarian, Eugene and Beurer-Kellner, Luca and Biggio, Battista and Ebert, Nico and Filip, David and Fischer, Marc and Frase, Heather and Hofer, David and Hoffmann, Juliane and Ippolito, Daphne and Jha, Somesh and McGregor, Sean and Mohammadi, Esfandiar and Nannini, Luca and Nita-Rotaru, Cristina and Oprea, Alina and Paeth, Kevin and Paverd, Andrew and Petit, Jonathan and Rauber, Andreas and Riess, Christian and Sotiropoulos, John and Wespi, Andreas and Grosse, Kathrin},
  year = {2026},
  month = sep,
  archiveprefix = {arXiv},
  primaryclass = {cs.CR},
}

Assessing Automated Prompt Injection Attacks in Agentic Environments

David Hofer, Edoardo Debenedetti, and Florian Tramèr

Jun 2026

Abs arXiv Bib PDF

Indirect prompt injection poses a critical threat to LLM agents that interact with untrusted external data, yet automated attack methods–proven effective for jailbreaking–remain underexplored in realistic agentic settings. We present a comprehensive empirical evaluation of automated prompt injection attacks against LLM agents, adapting both white-box (GCG) and black-box (TAP) methods to the agentic setting within the AgentDojo framework. We evaluate across 80 task pairs spanning four domains and multiple models, and find that black-box optimization substantially outperforms gradient-based methods, a gap we attribute to GCG’s optimization instability under reasonable compute budgets. We also find that TAP’s effectiveness depends on the attacker model, as both general capability and safety tuning affect attack success–stronger models produce more effective injections, while safety-tuned attackers can refuse to generate adversarial prompts. Task-universal attacks transfer effectively to unseen tasks and out-of-distribution domains, but attacks optimized on smaller open-source models do not transfer to frontier models like GPT-5. These findings highlight automated prompt injection as a credible but model-dependent threat, with significant barriers remaining for model-agnostic exploitation.
@misc{hofer2026assessing, title = {Assessing Automated Prompt Injection Attacks in Agentic Environments}, author = {Hofer, David and Debenedetti, Edoardo and Tram{\`e}r, Florian}, year = {2026}, month = jun, archiveprefix = {arXiv}, primaryclass = {cs.CR}, }
Automated Prompt Injection Attacks Against LLM Agents

David Hofer

MSc Thesis, ETH Zürich, Dec 2025

PDF

teaching

ETH Zürich, Department of Computer Science

Autumn 2026: AI Security Seminar
Autumn 2025: Privacy-Enhancing Technologies
Spring 2025: Machine Learning for Health Care
Autumn 2020: Linear Algebra