Causality in the Age of AI Scaling
5 May 2026
Tangier, Morocco
About the workshop
Reasoning about interventions, the core of causality, is fundamental to solving many of modern AI's most pressing challenges, including trustworthiness, reliability, explainability, and out-of-distribution generalization. Yet, recent AI breakthroughs have been overwhelmingly driven by scaling models on simple predictive objectives without explicit causal modeling, such as next-token prediction for Large Language Models or denoising prediction for diffusion models. This success raises a critical question for the community: Can causal abilities emerge from scale alone, and if not, what can explicit causal modeling bring that scale cannot? This workshop aims to understand this question and explore the potential synergy between scaling predictive methods and formal causal modeling to build the next generation of AI.
Workshop Goals
- Can causal abilities emerge from scale alone, and if not, what can explicit causal modeling bring that scale cannot?
- Is scaling sufficient for building intelligent systems? If not, is causal reasoning needed? What are the limitations and opportunities of scaling AI models?
- What are the challenges in developing scalable causal algorithms that can be potentially integrated into existing AI models?
- What ingredients are necessary to design robust, interactive world models?
Call for papers
We invite submissions exploring the synergy between scaling predictive methods and causal modeling to build the next generation of trustworthy and reliable AI.
Topics
Potential topics include, but are not limited to:
- Emergence of causal abilities in foundation models (or the failure thereof)
- OOD generalization and robustness of large models
- Scaling causal generative modeling and representation learning
- Causal, counterfactual, and logical reasoning in large models
- Design of interactive causal world models
- Trustworthy and interpretable AI
- Causal discovery and abstraction (especially applied to AI)
- Evaluation and benchmarking (and the limitations thereof)
Submission
We invite submissions of short papers presenting recent work on scaling and causality. Submissions are now being accepted through OpenReview.
Submissions should be formatted using the AISTATS LaTeX style. Papers are limited to 4 pages (excluding references and appendices). Accepted contributions will be presented as posters during the workshop. We will select a small number of contributed talks from the accepted submissions for short oral presentations at the workshop.
Submissions under review or accepted within the past year at other venues are allowed. All accepted papers are non-archival and will be made publicly available on OpenReview.
Important dates
- Submission deadline: February 27, 2026 (Anywhere on Earth)
- Notification of acceptance: March 18, 2026 (Anywhere on Earth)
- Workshop date: May 5, 2026
Keynote Speakers
Mihaela van der Schaar
University of Cambridge
Francesco Locatello
Institute of Science and Technology Austria
Sara Magliacane
University of Amsterdam
Alexander D'Amour
Google DeepMindSchedule
| Time | Activity |
|---|---|
| 9:00 - 9:10 | Welcome & Opening |
| 9:10 - 9:40 | Invited Talk 1 |
| 9:40 - 10:10 | Invited Talk 2 |
| 10:10 - 10:25 | Contributed Talks |
| 10:25 - 10:45 | Morning Break |
| 10:45 - 12:00 | Poster Session I |
| 12:00 - 1:00 | Lunch |
| 1:00 - 1:30 | Invited Talk 3 |
| 1:30 - 2:00 | Invited Talk 4 |
| 2:00 - 2:15 | Contributed Talks |
| 2:15 - 3:30 | Poster Session II |
| 3:30 - 4:00 | Invited Talk 5 |
| 4:00 - 4:55 | Panel Discussion |
| 4:55 - 5:00 | Closing Remarks |
Organizers
David Inouye
Purdue University
Bryon Aragam
University of Chicago