Belief as structural system, CAS self-correction, AI failure propagation

A proposed theoretical framework examines belief not as an individual conviction, but as a structural system that organizes meaning and legitimacy in human societies. This systems-thinking approach is also used to explore the conditions under which complex adaptive systems maintain self-correction. Another area of focus is how failures originating from AI systems can propagate through the technology supply chain, potentially leading to collective failures and large-scale societal harm.

A publication proposes that belief functions as a structural system within human societies. It suggests belief organizes meaning, legitimacy, and collective existence, moving beyond individual convictions. This framework uses Systems Theory to analyze belief as a relational architecture, interacting with symbolic systems, institutions, and other societal structures, investigating its role in establishing legitimacy and coordinating collective action.
An open science research program investigates the structural factors determining how complex adaptive systems preserve or lose their ability to self-correct. The program outlines a four-stage research structure, from theoretical concepts to empirical verification, and uses BenchEWS as a neutral framework for benchmarking.
A paper states that AI system failures or risks originating from one organization can extend across the AI technology supply chain and ecosystem. This propagation can result in widespread failures and societal harm.
A research article proposes a design for an advanced Network Monitoring System and Security Information and Event Management system. This system combines semantic knowledge representations with AI to improve context-aware incident management in complex IT environments, aiming to handle diverse data and offer detailed event context.
Shared Mobility Services are reshaping urban transportation by offering flexible travel options. These services can enhance travel times, decrease emissions, and reduce the number of vehicles on roads, contributing to system performance improvements.
A study applies two vulnerability assessment methods, the Livelihood Vulnerability Index (LVI) and the Climate Vulnerability Index (CVI), using a comprehensive perspective.
A study developed an indicator-based Resilience Capacity Index (RCI). This index serves to identify factors affecting the resilience capacity of surveyed households.
A paper introduces an architectural framework for analyzing AI safety, specifically for large language models. This framework incorporates concepts like Trustworthy AI, Responsible AI, and Ecosystemic Safe AI.
Researchers developed a framework for incorporating dissipation into stochastic dynamics. This framework balances dissipation with structure-preserving noise, maintaining the Gibbs measure on the symplectic manifold.
A document outlines an open science research program on the structural architecture of self-correction.
Research notes that synergies between Shared Mobility Services and Public Transport are often unaddressed in studies.
This research provides a review of multimodal transportation system models.
A multimodal traffic assignment model is introduced in this research.
Coordinated Motion Planning problems CMP-M and CMP-L were presented as a challenge at SoCG 2021.
A paper characterizes the sample complexity of Distributionally Robust Markov Decision Processes.
A study proposes strategies to increase household resilience in Bangladesh.
Researchers derived a stochastic counterpart to Euler-Lagrange equations.
Non-canonical stochastic Lagrangian/Hamiltonian systems were derived by researchers.
Researchers derived double-bracket dissipation using Lie group invariant stochastic dynamics.

Sources

The Architecture of Belief: A Theoretical Framework for Cultural Research
The Architecture of Belief: A Theoretical Framework for Cultural Research is a theoretical research publication that examines belief as one of the four fundamental architectures established within The Architecture of Humanity research program. Developed within the fields of artistic research and interdisciplinary humanities, the publication proposes that belief is not merely an individual conviction or a religious phenomenon, but a structural system through which human societies organize meaning, legitimacy, and collective existence. Drawing upon Systems Theory, Systemic Abstraction, artistic…
The Structural Architecture of Self-Correction: An Open Science Research Programme
The Structural Architecture of Self-Correction: An Open Science Research Programme ist das kanonische Einstiegsdokument eines offenen Forschungsprogramms zur Frage, unter welchen strukturellen Bedingungen komplexe adaptive Systeme ihre Fähigkeit zur Selbstkorrektur erhalten, stärken, anpassen, verschlechtern oder schließlich verlieren. Das Dokument beschreibt die zentrale wissenschaftliche Fragestellung, die vierstufige Forschungsarchitektur (Konzeption → Operationalisierung → mathematische Theorie → empirische Validierung), die Rolle von BenchEWS als neutralem Benchmarking-Framework sowie…
AI safety landscape for large language models: taxonomy, state-of-the-art, and future directions
AI safety is an emerging field of critical importance for the secure adoption and deployment of AI systems. With the recent advancements in large language models (LLMs), the technological landscape surrounding the design, development, and deployment of AI systems has undergone significant change. The failure of AI systems at one organization, or AI risks undertaken by one organization, can propagate down the AI technology supply chain, affect the entire AI ecosystem, and potentially lead to collective failures and cause large-scale harm to society. In this paper, we propose a novel…
Anomaly Detection using Knowledge Graphs: A Survey for Network Management and Cybersecurity Application
In telecommunications and computer networks, effective incident management highly depends on handling data heterogeneity and providing detailed event context. While knowledge graphs can assist with data integration and AI techniques with contextualization, these aspects are often treated separately, limiting progress toward detailed, explainable, shareable network behavior understanding. This article offers a structured overview, through three perspectives, of how integrating semantic knowledge representations and AI can address this gap. First, we analyze current Network Monitoring Systems…
User centric multimodal urban transportation network equilibrium including intermodality and shared mobility services
Shared Mobility Services (SMSs) are transforming urban transportation systems by offering flexible travel options. These services, which help reduce the number of cars on the roads, have the potential to enhance the transportation system’s performance, leading to improvements in travel times and emissions. This emphasizes the importance of assessing their impact on the system and users’ choices, particularly when integrated into complex multimodal systems that include public transport (PT). However, many studies overlook the synergies between SMSs and PT, leading to inaccurate traffic…
An assessment of the livelihood vulnerability of the riverbank erosion hazard and its impact on food security for rural households in Bangladesh
As the effects of climate change and hazards are starting to be felt worldwide, there are certain frontline countries that are most at risk and Bangladesh is genuinely at risk in terms of its economic viability and food security unless its citizens develop adaptation strategies to compensate for these effects. This study analyses how the impacts of climate change and hazards (specifically riverbank erosion) are already jeopardising the livelihood and food security of rural riparian (riverbank and char) households in Bangladesh, compromising their access to arable land, and thereby holding…
Semimartingale driven mechanics and reduction by symmetry for stochastic and dissipative dynamical systems
The recent interest in structure preserving stochastic Lagrangian and Hamiltonian systems raises questions regarding how such models are to be understood and the principles through which they are to be derived. By considering a mathematically sound extension of the Hamilton–Pontryagin principle, we derive a stochastic analog of the Euler–Lagrange equations, driven by independent semimartingales. Using this as a starting point, we can apply symmetry reduction carefully to derive non-canonical stochastic Lagrangian/Hamiltonian systems, including the stochastic Euler–Poincaré/Lie–Poisson…

Also this week

Full transcript

Belief is often treated as something personal, but what if it’s the structural system that organizes an entire society? That framework is our lead story on ComplexityPod, where we look at research through the lens of complex adaptive systems. We start there. A new theoretical publication is proposing that we think about belief in a different way—not just as a personal conviction, but as a kind of architecture that organizes entire societies. So, less like a private thought and more like a foundational system, something that provides the structure for meaning and collective action? Exactly. The paper calls it "The Architecture of Belief." It uses systems theory to look at how this structure interacts with other things, like power and identity, to guide a society. That idea of a foundational structure seems to connect with other research looking into the self-correction capabilities of complex adaptive systems. It does. An open science program has been started to identify the specific structural conditions that allow these systems to adapt and maintain themselves, or, conversely, what causes them to lose that ability. And it's a very structured investigation. The plan lays out four layers of research, moving from the purely conceptual all the way to empirical validation of the theories. This even drills down to the level of physics and mathematics. Some researchers have developed a framework to incorporate dissipation, or energy loss, into models of stochastic, or random, processes. But they're doing it in a very specific way. The goal is to ensure that the dissipation they introduce correctly balances out structure-preserving noise, so the overall dynamics of the system remain consistent with its underlying principles. These abstract ideas about complex systems have very concrete consequences when we look at engineered systems, especially artificial intelligence. A recent paper points out that failures in AI can now propagate from a single organization across the entire ecosystem. The proliferation of large language models means the risk one company takes can spread through the whole technology supply chain. Which can lead to collective failures and large-scale harm. To address this, that same paper proposes a new architectural framework for analyzing AI safety, integrating perspectives from Trustworthy AI, Responsible AI, and what it calls Ecosystemic Safe AI. That same challenge—managing complex technology—is also the focus of a design proposal for a next-generation incident management system for large IT environments. The idea is to integrate network monitoring with security event management. And to make it smarter. It would use semantic knowledge and AI to provide context-aware responses, tackling problems like data coming in from many different, incompatible sources through a mix of logic and probabilistic reasoning. We're seeing similar system-wide effects in cities, too. Shared Mobility Services are altering urban transportation networks by providing more flexible travel choices. And the potential effects are significant: improvements in system performance by reducing travel times and emissions, and even a reduction in the total number of cars on the road. This kind of holistic, systems-level approach is also being applied to social and environmental issues. A study in Bangladesh assessed the vulnerability of rural households to hazards like riverbank erosion. They used established methods like the Livelihood Vulnerability Index, but they also developed something new: an indicator-based Resilience Capacity Index. Which is a useful addition. It doesn't just measure vulnerability; it was created specifically to identify the factors that influence the resilience of those households, pointing toward strategies that could actually help enhance it. That's all for this issue. We'll be back next week with more research summaries. From ComplexityPod, thanks for listening.