Skip to content
Photo of Maxime Griot

Maxime Griot

Upcoming: AI Safety Research Fellow at Anthropic (November 2026)
Independent Researcher at ARISE, Jonathan Chen’s lab at Stanford University
PhD in Computer Science, UCLouvain (2026)
Doctor of Medicine (MD), UCLouvain (2023)

About

I work on AI safety and alignment: making frontier models honest about what they don’t know, robust to adversarial manipulation, and safe to deploy where mistakes are costly. I’m also a medical doctor, and medicine has been my testbed. During my PhD in Computer Science at UCLouvain, I showed that LLMs lack the metacognition needed for reliable medical reasoning and that multiple-choice benchmarks reward pattern recognition over medical knowledge. I also led the development of Internist-7B and of an Epic-integrated clinical agent now used by 3,900+ clinicians at the Cliniques Universitaires Saint-Luc.

After working on reasoning-model distillation at Sophont (March–May 2026), I joined ARISE, Jonathan Chen’s lab at Stanford, in August 2026 as an independent researcher, where I work on clinical data anonymization, synthetic data, and evaluations. In November 2026, I’m joining Anthropic as an AI safety research fellow. I’m especially interested in scalable oversight, honesty and calibration, adversarial robustness, and evaluations that measure what actually matters.

Highlighted publications view all →
AI safety & robustness
Large language models lack essential metacognition for reliable medical reasoning
paper
Clinical deployment
Implementation of large language models in electronic health records
paper
Evaluation
Pattern recognition or medical knowledge? The problem with multiple-choice questions in medicine
paper
Medical LLMs
Impact of high-quality, mixed-domain data on the performance of medical language models
paper

latest news

10/2026 I’ll be joining Anthropic in November 2026 as an AI Safety Research Fellow!
08/2026 Joined ARISE, Jonathan Chen’s lab at Stanford University, as an independent researcher, working on clinical data anonymization, synthetic data, and evaluations.
07/2026 Attacking Medical Vision-Language Models with Query-Based Zero-Order Optimization appears at the Trustworthy AI for Good (AI4GOOD) Workshop @ ICML 2026.
06/2026 New preprint with Paul Scotti and Tanishq Abraham — Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation.
05/2026 Presented Chopper, our clinical assistant integrated in Epic, to a visiting Japanese ministerial delegation at the Cliniques Universitaires Saint-Luc.
03/2026 Joined Sophont to work on distillation of reasoning models, through May 2026.

latest publications view all →

16publications7journal6conference & workshop2preprints1book chapter
  1. Attacking Medical Vision-Language Models with Query-Based Zero-Order Optimization
    M. Griot
    ICML 2026 Workshop Trustworthy AI for Good (AI4GOOD) Workshop @ ICML 2026 2026
    paper

    Griot, M. (2026). "Attacking Medical Vision-Language Models with Query-Based Zero-Order Optimization." Trustworthy AI for Good (AI4GOOD) Workshop @ ICML 2026.

  2. Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation
    M. Griot, P. S. Scotti, and T. M. Abraham
    arXiv arXiv preprint arXiv:2606.05988 2026
    paper

    Griot, M., Scotti, P. S., & Abraham, T. M. (2026). "Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation." arXiv preprint arXiv:2606.05988.

  3. torchtune: PyTorch native post-training library
    M. Obozov, M. Griot, J. Cummings, E. Smothers, F. Mello, R. Ayub, P. J. Bontrager, S. Mohammadi, A. Kwiatkowski, N. Azrak, and M. Mironenco
    arXiv arXiv preprint arXiv:2605.21442 2026
    papercode

    Obozov, M., Griot, M., Cummings, J., Smothers, E., Mello, F., Ayub, R., Bontrager, P. J., Mohammadi, S., Kwiatkowski, A., Azrak, N., & Mironenco, M. (2026). "torchtune: PyTorch native post-training library." arXiv preprint arXiv:2605.21442.

  4. Medmarks: An Open-Source LLM Benchmark Suite for Medical Tasks
    B. Warner, R. S. Grandhi, M. Kieffer, …, M. Griot, …, P. S. Scotti, and T. M. Abraham (35 authors)
    ICML 2026 Workshop ICML 2026 Workshop on Multi-modal Foundation Models and Large Language Models for Life Sciences 2026
    paper

    Warner, B., Grandhi, R. S., Kieffer, M., et al., Griot, M., et al., Scotti, P. S., & Abraham, T. M. (2026). "Medmarks: An Open-Source LLM Benchmark Suite for Medical Tasks." ICML 2026 Workshop on Multi-modal Foundation Models and Large Language Models for Life Sciences.

  5. Implémentation d'un chatbot dans le dossier patient informatisé
    M. Griot, A. Irrthum, J. Vanderdonckt, and D. Yuksel
    Proceedings Actes de la journée d'étude sur l'utilisation des LLM à l'hôpital 2026
    paper

    Griot, M., Irrthum, A., Vanderdonckt, J., & Yuksel, D. (2026). "Implémentation d'un chatbot dans le dossier patient informatisé." Actes de la journée d'étude sur l'utilisation des LLM à l'hôpital.