March 2025 Newsletter ❇️🐞🌗 Active Inference Institute

See our first quarterly Roundtable of the year from March 28th, giving overview of this year’s activities and directions in livestreamed form.

Here is the entry point for 2025, giving ways to get involved with ongoing activities at the Institute and Ecosystem this year.

Updates from Institute Projects

• The Scientific Advisory Board (SAB) for 2025 has had meetings in February and March.
• Institute Officers applied for funding, and did not receive it, for “Increasing the Accessibility and Applicability of Active Inference: Generative Playbooks and Open-Source Summer School Curriculum Development” (published as preprint). If you are interested in (contributing to, funding, supporting) either of these projects, get in touch.
• The Theoretical Neurobiology (TNB) Group meetings continue. Information on how to join these weekly meetings can be found at the project page.
• Some of highlight livestreams from the last month included:
• ModelStream #017.1 ~ What's Wrong with Active Inference and How Lazy Dynamics Is Set to Solve It (Lazy Dynamics and guests)
• GuestStream #099.1 ~ Avel Guénin-Carlut: “From Contextuality to Social Constraints — An Active Inference ontology for participatory realism and social change”
• GuestStream #100.1 ~ Yu Kanazawa: The free energy principle and its implications to language learning and education
• The Textbook Group reports: Activity continues for Cohorts 7 and 8. Registration is open.
• We are continuing development on a possible Active Inference Seasonal School, to be held in-person later in 2025 or in 2026. Please reach out if you are interested in being a physical host, facilitator, sponsor, or other partner for this effort.

Updates from Ecosystem Projects

• The Research Discovery Engine project (GEN-25) reports: We’re building a platform that leverages advanced AI, including LLMs, Topic Modeling, Agents and Active Inference to accelerate and increase the impact of scientific research. This will be available on Mobile and the web (work in progress website). We applied for funding, launched the website, created a Github, produced a product roadmap, began integrating several functional database and AI oriented back end services, and began collaborating in real time and asynchronously. Team: Andrea Farias, Andrew Pashea, Austin Cook, Benedikt Waldeck, Daniel Friedman, Janna Lumiruusu, Marcus Appelros, Shagor Rahman, Vladimir Baulin.
• Kolja J. Kurzer reports: The prototype of the idea developed in my master thesis finally materialized into a website. I built the website using Obsidian and Quartz 4 and hosted it via Github pages. Most of the included theoretical sentences included thus far are from Millidge et al. 2021 “A Mathematical Walkthrough and Discussion of the Free Energy Principle”. I just arrived at EFE and am planning to include the low and high road from the Parr et al. Textbook next. Launch of the website. First 40 Sentences. First presentation about the idea and the prototype 21.03.2025 in Leipzig, Germany.
• Jonathan Shock reports: we are currently advertising for two postdocs working with myself and Mark Solms on our affective Active Inference project. See [quarterly roundtable for full call].
• From the “Humanity’s Story of an Uncertain Self” project, Shagor Rahman (Discord: Shaggstar, or email) reports: Formalized the Myth of Objectivity Hypothesis.
• Harshil Shah, Satyaki Maitra, Rohit Shenoy on the project “An Active Inference Approach to Autonomous Navigation” report: We have completed phase 1 of our project, submitting to our local science fair and winning 2nd place in our category. We havecompleted a poster summarizing our project below, and aGithub showing the code so far. We learned how to present active inference to people using examples and technical mathematics. We applied Active Inference to autonomous navigation in a short period of time. The work expands active inference into field of navigation in hardware (future goals). We accessed numerous papers and help from people within the Institute. We are asking now: How can we move into more realistic simulations and hardware with active inference?

• Do you have any updates from your projects or research from the Active Inference Ecosystem? Complete a measurement form and we may include it in a future newsletter.

Support and Engage

Donate to the Institute to support our work and sustainability in the years to come.

Subscribe to or share the newsletter, or contact us with other philanthropic and grant ideas. The Affordances page lists targeted opportunities for volunteer contributions.

Get in touch if you have general or specific feedback, ideas, or questions for the Institute.

Officers