Generative Artificial Intelligence (GenAI) is revolutionising education, but in doing so, it is often seen simply as a tool for automation and efficiency. This paper challenges this narrow perspective, arguing that GenAI’s transformative potential lies in fostering Meta-AI skills—specialised metacognitive competencies for engaging AI as a cognitive partner. Our objective is to explore how GenAI reshapes learning environments and necessitates a pedagogical shift, hypothesising that traditional education paradigms are insufficient without Meta-AI skills to navigate GenAI’s dynamic outputs. Drawing on constructionist theory, we conceptually analyse seven phenomena: learners’ perceptions, reasoning transitions, scientific research paradigms, information processing shifts, tacit knowledge articulation, multimodal interactions, and early AI education.. The methodology involves synthesising theoretical and empirical literature to frame Meta-AI skills as essential for modern learning. Key results reveal that GenAI transforms education into interactive, exploratory spaces, shifting from symbolic to indexical processing and generalisation-based to situated reasoning. Multimodal GenAI enhances metacognitive awareness, while tacit knowledge exploration deepens self-reflection—outcomes requiring Meta-AI skills beyond conventional metacognition. Practically, educators can integrate these skills into curricula through scaffolded prompt engineering (e.g., refining AI queries), multimodal projects (e.g., creating cross-media narratives), and critical evaluation of AI outputs, empowering students to leverage GenAI effectively. We conclude that GenAI’s integration demands a reevaluation of pedagogy, with Meta-AI skills bridging theoretical shifts and practical applications.

This paper extends Rodney Brooks’ subsumption architecture into the era of Agentic AI by replacing its priority arbiter with a Generative Orchestrator that performs semantic mediation—interpreting heterogeneous agent outputs and integrating them into a coherent action rather than merely arbitrating among them. Brooks’ original model (1986) demonstrated that autonomous behavior can emerge from parallel reactive layers without symbolic representation, establishing principles later recognized as foundational to agentic systems: environmental responsiveness, autonomy, and goal-directed action. Contemporary Agentic AI, however, requires capabilities beyond mechanical response—decision-making, adaptive strategy, and goal pursuit. We therefore reinterpret subsumption layers as four interacting agent types: reflex, model-based, goal-based, and utility-based, coordinated through semantic mediation. The Generative Orchestrator employs large language models not for content generation but for decision synthesis, enabling integrative agentic behavior. This approach merges real-time responsiveness with interpretive capacity for learning, reasoning, and explanation. An autonomous driving case study demonstrates how the architecture sustains behavioral autonomy while generating human-interpretable rationales for its actions. Validation was conducted through a Python-based proof-of-concept on an NVIDIA platform, reproducing the scenario to evaluate and confirm the architecture. This framework delineates a practical pathway toward advancing autonomous agents from reactive control to fully Agentic AI systems capable of operating in open, uncertain environments.

Multi-agent perception systems must operate under fundamental asymmetries: some agents provide fast but unreliable observations, while others deliver higher-quality evidence with delay and uncertain correspondence. Traditional deterministic orchestration and rule-based fusion struggle to manage these trade-offs, often producing brittle or unstable behaviour. We introduce a probabilistic orchestration framework that treats coordination as an epistemic generation problem —constructing and updating belief states under uncertainty—rather than a selection problem. Instead of committing to a single agent’s output, the orchestrator constructs a belief state that explicitly represents uncertainty, evidential provenance, and temporal relevance. Decisions are produced through latency-aware, association-weighted fusion, and uncertainty itself becomes a first-class signal governing action, deferral, and learning. Crucially, the orchestrator enables controlled teacher–student adaptation: high-confidence, well-associated stationary observations are gated into a feedback loop that improves ego perception over time while mitigating error amplification. We demonstrate the approach on an infrastructure-assisted dual-camera obstacle-recognition task. Experimental results show improved robustness to distance, occlusion, and delayed evidence compared to ego-only and deterministic orchestration baselines. By operationalizing orchestration as epistemic generation, this work provides a unifying framework for robust decision-making and safe adaptation in multi-agent systems, with implications that extend beyond perception to agentic and generative AI architectures.

Multi-agent perception systems must operate under fundamental asymmetries: some agents provide fast but unreliable observations, while others deliver higher-quality evidence with delay and uncertain correspondence. Traditional deterministic orchestration and rule-based fusion struggle to manage these trade-offs, often producing brittle or unstable behavior. We introduce a probabilistic orchestration framework that treats coordination as an epistemic generation problem—constructing and updating belief states under uncertainty—rather than a selection problem. Instead of committing to a single agent’s output, the orchestrator constructs a belief state that explicitly represents uncertainty, evidential provenance, and temporal relevance. Decisions are produced through latency-aware, association-weighted fusion, and uncertainty itself becomes a first-class signal governing action, deferral, and learning. Crucially, the orchestrator enables controlled teacher–student adaptation: high-confidence, well-associated stationary observations are gated into a feedback loop that improves ego perception over time while mitigating error amplification. We demonstrate the approach on an infrastructure-assisted dual-camera obstacle-recognition task. Experimental results show improved robustness to distance, occlusion, and delayed evidence compared to ego-only and deterministic orchestration baselines. By operationalizing orchestration as epistemic generation, this work provides a unifying framework for robust decision-making and safe adaptation in multi-agent systems, with implications that extend beyond perception to agentic and generative AI architectures.