About Me

I am a researcher focused on the integration of statistical inference, causal representational learning, mechanistic models, and probabilistic modeling with deep generative multimodal models. I work as a Research Scientist at Microsoft Research, and I have previously held roles across big tech and startups as an AI researcher, research engineer, and data scientist.

I am the author of the book Causal AI (• Amazon • Publisher), which introduces causal reasoning for machine learning practitioners through a practical, intuitive lens. I am also a core contributor to several open-source projects, including GraphRAG, PyWhy, and bnlearn.

You can find more of my publications and professional details on
Google Scholar and LinkedIn.

---

Selected Publications

• Multiverse Mechanica: A testbed for learning game mechanics via counterfactual Worlds
  R. Ness, R. Cannizzaro, Y. Wu, L. Kunze – 2025
  Introduces a counterfactual generative testbed for evaluating causal world models and game mechanics.

• Causal Reasoning and Large Language Models: Opening a new frontier for causality E. Kiciman, R. Ness, A. Sharma, C. Tan – Transactions on Machine Learning Research, 2023
  An exploration of how large language models intersect with classical causal inference frameworks.

• Walk the talk? Measuring the faithfulness of large language model explanations
  K. Matton, R. O. Ness, J. Guttag, E. Kıcıman – arXiv preprint arXiv:2504.14150, 2025
  Investigates the alignment between LLM explanations and the true underlying model behavior.

• A Bayesian active learning experimental design for inferring signaling networks
  R. O. Ness, K. Sachs, P. Mallick, O. Vitek – Journal of Computational Biology, 2018
  Presents Bayesian experimental design methods for inferring biological signaling pathways.

---

Notable Collaborations

• From Local to Global: A Graph RAG approach to query-focused summarization
  D. Edge, H. Trinh, N. Cheng, J. Bradley, A. Chao, A. Mody, … R. O. Ness – arXiv preprint arXiv:2404.16130, 2024
  A scalable approach to retrieving and summarizing large corpora using graph-structured retrieval-augmented generation.

• Can generalist foundation models outcompete special-purpose tuning? A case study in medicine
  H. Nori, Y. T. Lee, S. Zhang, D. Carignan, R. Edgar, N. Fusi, N. King, J. Larson, Y. Li, W. Liu, R. Luo, S. M. McKinney, R. Osazuwa Ness, H. Poon, T. Qin, N. Usuyama, C. White, E. Horvitz – 2023
  Evaluates generalist versus specialist foundation models on a range of medical tasks and benchmarks.

• Generative Propaganda
  M. I. G. Daepp, A. Cuevas, R. O. Ness, V. Y.-P. Wang, B. K. Nayak, D. Mishra, T.-C. Cheng, S. Desai, J. Pal – arXiv preprint arXiv:2509.19147, 2025
  A long-form collaborative study of how generative models can shape propaganda, disinformation, and media ecosystems.

---

Media, Talks & Interviews

• Video will kill the truth if monitoring doesn’t improve, argue two researchers
  By-invitation article in The Economist, co-authored with Madeleine Daepp – March 26, 2024

• Learning Bayesian Statistics Podcast – Episode #137: Causal AI & Generative Models
  July 23, 2025

• Super Data Science Podcast – Episode #909: Causal AI with Robert Osazuwa Ness
  July 29, 2025

• Hockey Stick Show #26 – From Zombie Fungus to Causal AI October 07, 2024

• Causal Bandits Podcast #11: From Biology to Generative AI & RL | Causal Models with Robert Ness
  November 12, 2023

• TWIML Podcast – Episode #638: Are LLMs Good at Causal Reasoning? July 14, 2023

• BP Causal Inference Symposium 2023

• ODSC Workshop 2021 – Causal AI March 21, 2021

• TWIML Podcast – Episode #342: Causality 101 with Robert Ness January 27, 2020

---