DX space

Participant Registration Open: 4th Applied Active Inference Symposium on November 13-15, 2024 https://activeinferenceinstitute.substack.com/p/participant-registration-open-4th

December 2024 Newsletter 🎹🎡💾 Active Inference Institute https://open.substack.com/pub/activeinferenceinstitute/p/december-2024-newsletter-active-inference https://newsletter.activeinference.institute/

春节快乐！ https://youtu.be/XUaYym2v-BQ

New publication: "The Ant Stack" https://zenodo.org/records/16782757 We present the Ant Stack: a compact, modular framework that emulates an ant from physics to cognition. It comprises three interoperable layers— AntBody (morphology, actuation, sensors), AntBrain (functional neural circuits templated from conserved insect architectures), and AntMind (active inference and stigmergy for individual and collective cognition). The stack uses explicit I/O contracts, sparse/low‑power compute, and species‑parameterized configurations aligned with current myrmecology.

well, that escalated quickly

Blossom by blossom the Active Inference spring begins. https://resiliencemedia.co/stanhope-ai-raises-8m-for-new-approach-of-ai-for-physical-applications/

ActInf Inside™ https://youtu.be/yRV8fSw6HaE

Active Inference for Physical AI Agents - An Engineering Perspective Bert de Vries, 21 Mar 2026 "Physical AI agents, such as robots and other embodied systems operating under tight and fluctuating resource constraints, remain far less capable than biological agents in open-ended real-world environments. This paper argues that Active Inference (AIF), grounded in the Free Energy Principle, offers a principled foundation for closing that gap. We develop this argument from first principles, following a chain from probability theory through Bayesian machine learning and variational inference to active inference and reactive message passing. From the FEP perspective, systems that maintain their structural and functional integrity over time can, under suitable assumptions, be described as minimizing variational free energy (VFE), and AIF operationalizes this by unifying perception, learning, planning, and control within a single computational objective. We show that VFE minimization is naturally realized by reactive message passing on factor graphs, where inference emerges from local, parallel computations. This realization is well matched to the constraints of physical operation, including hard deadlines, asynchronous data, fluctuating power budgets, and changing environments. Because reactive message passing is event-driven, interruptible, and locally adaptable, performance degrades gracefully under reduced resources while model structure can adjust online. We further show that, under suitable coupling and coarse-graining conditions, coupled AIF agents can be described as higher-level AIF agents, yielding a homogeneous architecture based on the same message-passing primitive across scales. Our contribution is not empirical benchmarking, but a clear theoretical and architectural case for the engineering community." https://arxiv.org/abs/2603.20927

Yet another sharp team applying active inference has come together to disrupt robotics. If scouting early-stage deep-tech is your job, you should reach out to them today. https://theblanket.ai/

Two brains are better than one https://med.stanford.edu/news/all-news/2026/09/two-separate-brains.html