When Agentic AI started dominating research papers, demos, and conference talks, I was curious but cautious. The idea of intelligent agents, autonomous systems powered by large language models that can plan, reason, and take actions using tools, sounded brilliant in theory. But I wanted to know what happened when you used them. Not in a toy notebook or a slick demo, but in real projects, with real constraints, where things needed to work reliably and repeatably. In my role as Clinical AI & Data Scientist at Bayezian Limited, I work at the intersection of data science, statistical modelling, and clinical AI governance, with a strong emphasis on regulatory-aligned standards such as CDISC. I have been directly involved in deploying agentic systems into environments where trust and reproducibility are not optional. These include real-time protocol compliance, CDISC mapping, and regulatory workflows. We gave agents real jobs. We let them loose on messy documents. And then we watched them work, fail, learn, and (sometimes) recover. This article is not a critique of Agentic AI as a concept.

We need a new ethics for a world of AI agents . The deployment of capable AI agents raises fresh questions about safety, human-machine relationships and social coordination. Ar tificial intelligence (AI) developers are shifting their focus to building agents that can operate independently, with little human inter vention. To be an agent is to have the ability to perceive and act on an environment in a goal-directed and autonomous way. For example, a dig ital agent could be programmed to browse the web and make online purchases on behalf of a user -- comparing prices, selecting items and completing checkouts.

The goal of Multi-Agent Path Finding (MAPF) is to find a set of paths for a fleet of agents moving in a shared environment such that the agents reach their goals without colliding with each other. In practice, some of the robots executing the plan may get delayed, which can introduce collision risk. Although robust execution methods are used to ensure safety even in the presence of delays, the delays may still have a significant impact on the duration of the execution. At some point, the accumulated delays may become significant enough that instead of continuing with the execution of the original plan, even if it was optimal, there may now exist an alternate plan which will lead to a shorter execution. However, the problem is how to decide when to search for the alternate plan, since it is a costly procedure. In this paper, we propose a holistic architecture for robust execution of MAPF plans, its monitoring and optimization. We exploit a robust execution method called Action Dependency Graph to maintain an estimate of the expected execution duration during the plan's execution. This estimate is used to predict the potential that finding an alternate plan would lead to shorter execution. We empirically evaluate the architecture in experiments in a real-time simulator which we designed to mimic our real-life demonstrator of an autonomous warehouse robotic fleet.

Abstract--As sixth-generation (6G) networks move toward ultra-dense, intelligent edge environments, efficient resource management under stringent privacy, mobility, and energy constraints becomes critical. This paper introduces a novel Federated Multi-Agent Reinforcement Learning (Fed-MARL) framework that incorporates cross-layer orchestration of both the MAC layer and application layer for energy-efficient, privacy-preserving, and real-time resource management across heterogeneous edge devices. Each agent uses a Deep Recurrent Q-Network (DRQN) to learn decentralized policies for task offloading, spectrum access, and CPU energy adaptation based on local observations (e.g., queue length, energy, CPU usage, and mobility). T o protect privacy, we introduce a secure aggregation protocol based on elliptic-curve Diffie-Hellman key exchange, which ensures accurate model updates without exposing raw data to semi-honest adversaries. We formulate the resource management problem as a partially observable multi-agent Markov decision process (POMMDP) with a multi-objective reward function that jointly optimizes latency, energy efficiency, spectral efficiency, fairness, and reliability under 6G-specific service requirements such as URLLC, eMBB, and mMTC. Simulation results demonstrate that Fed-MARL outperforms centralized MARL and heuristic baselines in task success rate, latency, energy efficiency, and fairness, while ensuring robust privacy protection and scalability in dynamic, resource-constrained 6G edge networks. Sixth-generation (6G) wireless networks are poised to transform communication systems by enabling ultra-dense connectivity, low-latency services, and intelligent edge processing capabilities [1]. These advances are critical for emerging applications such as autonomous driving, augmented reality, and massive Internet of Things (IoT) deployments, each imposing diverse and stringent quality-of-service (QoS) requirements [2], [3]. Efficiently meeting these demands requires decentralized, real-time resource management frameworks capable of operating in highly dynamic, interference-prone, and energy-constrained environments under strict privacy conditions. Traditional centralized resource management architectures, which depend on global network knowledge for task offload-ing, spectrum allocation, and computational scheduling, face significant limitations in 6G contexts [4], [5]. These include scalability bottlenecks, latency, communication overhead, and privacy concerns, particularly when raw user data must be aggregated [6].

The convergence of Web3 technologies and AI agents represents a rapidly evolving frontier poised to reshape decentralized ecosystems. This paper presents the first and most comprehensive analysis of the intersection between Web3 and AI agents, examining five critical dimensions: landscape, economics, governance, security, and trust mechanisms. Through an analysis of 133 existing projects, we first develop a taxonomy and systematically map the current market landscape (RQ1), identifying distinct patterns in project distribution and capitalization. Building upon these findings, we further investigate four key integrations: (1) the role of AI agents in participating in and optimizing decentralized finance (RQ2); (2) their contribution to enhancing Web3 governance mechanisms (RQ3); (3) their capacity to strengthen Web3 security via intelligent vulnerability detection and automated smart contract auditing (RQ4); and (4) the establishment of robust reliability frameworks for AI agent operations leveraging Web3's inherent trust infrastructure (RQ5). By synthesizing these dimensions, we identify key integration patterns, highlight foundational challenges related to scalability, security, and ethics, and outline critical considerations for future research toward building robust, intelligent, and trustworthy decentralized systems with effective AI agent interactions.

As Cellular Vehicle-to-Everything (C-V2X) evolves towards future sixth-generation (6G) networks, Connected Autonomous Vehicles (CAVs) are emerging to become a key application. Leveraging data-driven Machine Learning (ML), especially Deep Reinforcement Learning (DRL), is expected to significantly enhance CAV decision-making in both vehicle control and V2X communication under uncertainty. These two decision-making processes are closely intertwined, with the value of information (VoI) acting as a crucial bridge between them. In this paper, we introduce Sequential Stochastic Decision Process (SSDP) models to define and assess VoI, demonstrating their application in optimizing communication systems for CAVs. Specifically, we formally define the SSDP model and demonstrate that the MDP model is a special case of it. The SSDP model offers a key advantage by explicitly representing the set of information that can enhance decision-making when available. Furthermore, as current research on VoI remains fragmented, we propose a systematic VoI modeling framework grounded in the MDP, Reinforcement Learning (RL) and Optimal Control theories. We define different categories of VoI and discuss their corresponding estimation methods. Finally, we present a structured approach to leverage the various VoI metrics for optimizing the ``When", ``What", and ``How" to communicate problems. For this purpose, SSDP models are formulated with VoI-associated reward functions derived from VoI-based optimization objectives. While we use a simple vehicle-following control problem to illustrate the proposed methodology, it holds significant potential to facilitate the joint optimization of stochastic, sequential control and communication decisions in a wide range of networked control systems.

Large Language Models (LLMs) are increasingly used to build autonomous agents that perform complex tasks with external tools, often exposed through APIs in enterprise systems. Direct use of these APIs is difficult due to the complex input schema and verbose responses. Current benchmarks overlook these challenges, leaving a gap in assessing API readiness for agent-driven automation. We present a testing framework that systematically evaluates enterprise APIs when wrapped as Python tools for LLM-based agents. The framework generates data-aware test cases, translates them into natural language instructions, and evaluates whether agents can correctly invoke the tool, handle their inputs, and process its responses. We apply the framework to generate over 2400 test cases across different domains and develop a taxonomy of common errors, including input misinterpretation, output failures, and schema mismatches. We further classify errors to support debugging and tool refinement. Our framework provides a systematic approach to enabling enterprise APIs as reliable tools for agent-based applications.

We present a machine learning framework for adaptive source localization in which agents use a structure-preserving digital twin of a coupled hydrodynamic-transport system for real-time trajectory planning and data assimilation. The twin is constructed with conditional neural Whitney forms (CNWF), coupling the numerical guarantees of finite element exterior calculus (FEEC) with transformer-based operator learning. The resulting model preserves discrete conservation, and adapts in real time to streaming sensor data. It employs a conditional attention mechanism to identify: a reduced Whitney-form basis; reduced integral balance equations; and a source field, each compatible with given sensor measurements. The induced reduced-order environmental model retains the stability and consistency of standard finite-element simulation, yielding a physically realizable, regular mapping from sensor data to the source field. We propose a staggered scheme that alternates between evaluating the digital twin and applying Lloyd's algorithm to guide sensor placement, with analysis providing conditions for monotone improvement of a coverage functional. Using the predicted source field as an importance function within an optimal-recovery scheme, we demonstrate recovery of point sources under continuity assumptions, highlighting the role of regularity as a sufficient condition for localization. Experimental comparisons with physics-agnostic transformer architectures show improved accuracy in complex geometries when physical constraints are enforced, indicating that structure preservation provides an effective inductive bias for source identification.

The work describes the development of a hybrid control architecture for an anthropomorphic tour guide robot, combining a multi-agent resource management system with automatic behavior scenario generation based on large language models. The proposed approach aims to overcome the limitations of traditional systems, which rely on manual tuning of behavior scenarios. These limitations include manual configuration, low flexibility, and lack of naturalness in robot behavior. The process of preparing tour scenarios is implemented through a two-stage generation: first, a stylized narrative is created, then non-verbal action tags are integrated into the text. The multi-agent system ensures coordination and conflict resolution during the execution of parallel actions, as well as maintaining default behavior after the completion of main operations, contributing to more natural robot behavior. The results obtained from the trial demonstrate the potential of the proposed approach for automating and scaling social robot control systems.

Automated characterization of porous materials has the potential to accelerate materials discovery, but it remains limited by the complexity of simulation setup and force field selection. We propose a multi-agent framework in which LLM-based agents can autonomously understand a characterization task, plan appropriate simulations, assemble relevant force fields, execute them and interpret their results to guide subsequent steps. As a first step toward this vision, we present a multi-agent system for literature-informed force field extraction and automated RASPA simulation setup. Initial evaluations demonstrate high correctness and reproducibility, highlighting this approach's potential to enable fully autonomous, scalable materials characterization.