April 2026 Newsletter 🌿 🧅 🛠️ Active Inference Institute
Hello all, we have many updates to share this month, and much more upcoming for this year.
The Fundamentals of Active Inference textbook group kicked off the first meeting on April 21st. It’s not too late to join the FUN-damentals. All details and registration form here. The textbook is “Fundamentals of Active Inference: Principles, Algorithms, and Applications of the Free Energy Principle for Engineers” by Sanjeev V. Namjoshi (2026).
Abstract registration is open until May 24th for the 7th International Workshop on Active Inference (IWAI). The workshop will take place in the Spanish National Research Council, Madrid, Spain, from 14-16 October, 2026. https://iwaiworkshop.github.io/
Updates from Institute Fellows
Shagor Rahman reports a new publication with Andrew Pashea in Frontiers in Sociology, “Transcendental model selection: a computational account of symbolic cognition and general intelligence through morality and culture”. They wrote, “This framework can be evaluated through simulations similar to recent active inference literature and provides a foundation for building generally intelligent systems aligned with human cultural values.”
Jean-Francois Cloutier in the Symbolic Cognitive Robotics projects writes: I am making daily, steady progress in designing and implementing a cognitive robotics architecture meant to give a Lego robot free agency. I see three dimensions to its (predictive, integrative, sense-making, normative, hierarchical, distributed) cognition: perception, action and reflection. I completed a first pass through perception and I am currently focusing on action. I have a detailed design for how a dynamic, multi-actor society of mind decides what to do next to “feel” better and how it carries out its intents. I am coding to this design and iterating on it as limitations come into view.
Updates from Projects
Daniel Derome reports a new publication: “Preventing a Civilizational Collapse With The Help of The Free Energy Principle”. He writes: My goal is to create an organization dedicated to preventing a civilizational collapse, together with anyone who believes that our conceptions of the world can converge. I am aware that this project seems utopian at first glance. That is why I hasten to add that the scale of the action would be intergenerational, i.e., we should not expect such an organization to significantly transform the state of the world in the short term.Ultimately, the problem we need to solve is one of scales: the collective scale, the scale of History, and the scale of conception of the world.
The Artificial Sentience project (Cleber Gomes) reports: a new blog post on Fear as the Counterpart of Motivation: A Computational ExplorationThis post explores fear as the computational counterpart of motivation: a process not of energizing action, but of constraining it. While motivation pulls the agent toward goals, fear shapes the space of actions that are considered safe.
Daniel Friedman and Joel Dietz report a new publication “A Living Meta-Analysis Architecture for Active Inference: Assertion Extraction, Nanopublications, and Hypothesis Scoring”. This project employs multi-source retrieval, LLM-based assertion extraction (Nanopublications), probabilistic knowledge graphs, and citation-weighted hypothesis scoring to map the structural and thematic trajectory of our field! All methods are open source at https://github.com/ActiveInferenceInstitute/act_inf_metaanalysis/
Daniel Friedman provides a new publication: “Towards Lean 4 Formalization of the Free Energy Principle: AI-Driven Theorem Sketching and Verification for Active Inference and Bayesian Mechanics”. The net contribution and current direction of the work is to convert long-standing qualitative arguments about the mathematical status of the FEP into a maintainable, version-pinned, publicly auditable record of exactly which claims are machine-checkable and which remain open; methods, source, catalog, figures. The end-to-end reproduction of this manuscript are released at the FEP_Lean open-source repository https://github.com/ActiveInferenceInstitute/FEP_Lean via the template approach https://github.com/docxology/template which injects updated validated package-level and manuscript-level metadata into the template as rendered and versioned in practice.
Some highlight livestreams from the last month included:
ModelStream #007.3 with Conor Heins, Arun Niranjan, Dmitrije Markovic, Peter Thestrup Waade, Paul Kinghorn, Jonas Mago on pymdp 1.0.0.
If you would like to help make audio-visual content with the Institute, there are many roles available, for example to help plan livestreams. See this video for more information.
This page gives a 1-page summary of the structure of the Institute — a helpful entry point for those who are new, or learning about our (growing) organizational morphology. Donate to the Institute to support our impact and sustainability.
Here is the entry point for 2026, and ways to get involved with the Institute and Ecosystem this year. Contact us with other philanthropic and grant ideas.
Get in touch if you have general or specific feedback, ideas, or questions for the Institute.
Complete this form to have your research, education, and work updates included in a future newsletter.
More to come! Stay tuned via the newsletter and Discord.
Both with Certainty and Uncertainty,
Officers, Alexandra & Daniel
👣👣
