Agentic AI systems that reason, plan, and act on complex goals have advanced rapidly across software engineering, scientific discovery, drug development, healthcare, finance, and social simulation. Across these domains a single failure pattern recurs: current systems can execute tasks competently but often struggle to determine when to act, when to pause, when to change strategy, and when to involve a human. Existing reviews catalog agentic architectures, taxonomies, and limitations, but none specify what capabilities these systems must acquire to support dynamic human-AI collaboration. We address that gap. We define collaborative AI as a class of systems that combine generative exploration with autonomous action and calibrate between them based on context, uncertainty, and task demands. We identify four required capabilities: metacognition, contextual mode-switching, uncertainty-aware action, and adaptive human collaboration. We relate these capabilities to established multi-agent systems foundations, including belief-desire-intention architectures, adjustable autonomy, mixed-initiative interaction, and decentralized decision-theoretic control, while specifying the distinct challenges that LLM-based agents introduce. Across the six domains reviewed here, these gaps appear repeatedly and are not solved by current architectures, which positions collaborative AI as a concrete near-term research objective.
Nalan Karunanayake, Savindu Nanayakkara, Kasun Gayashan Hettihewa et al.· International Journal of Net...· 0 citations
Groundwater is vital to subsurface ecosystems and to maintaining water supplies for human and environmental needs. Heavy metal pollution of water bodies poses a significant threat to environmental health and human well-being. In this paper, a detailed spatiotemporal analysis of heavy metal pollution at a network of 35 sampling sites is presented for the summer, autumn, and winter seasons. The water samples were analyzed based on the concentration (μg/L) of eight priority metals, such as Lead (Pb), Mercury (Hg), Cadmium (Cd), Arsenic (As), Chromium VI (Cr-VI), Copper (Cu), Zinc (Zn), as well as Iron (Fe). Measuring contamination levels, identifying space hotspots, and explaining seasonal variations were the key tasks. The results show that Fe, Zn, and Cu concentrations are consistently high across seasons, with mean values ranging from 234.3 to 253.2 µg/L (Fe), 163.2 to 175.6 µg/L (Zn), and 107.0 to 109.2 µg/L (Cu), indicating a widespread geogenic or diffuse source. The most significant seasonal deviation was observed in the fall when there were unusually high levels of Cd and Hg, with mean concentrations reaching 0.476 µg/L and 0.312 µg/L, respectively, suggesting a strong seasonal contamination event or mobilization process. The spatial analysis showed that the locations (e.g., L6, L8, L10, L28, L29) exhibited common hotspots for different metals, with maximum concentrations reaching up to 793.3 µg/L (Fe), 602.3 µg/L (Zn), and 508 µg/L (Cu). Principal Component Analysis (PCA) effectively separated seasonal trends and classified metals into anthropogenic (Pb, Cd, Hg, Cr (VI)) and geogenic/diffuse (Fe, Zn, Cu) groups. The health risk assessment indicated no significant non-carcinogenic risk, although children are more vulnerable, while arsenic levels in winter approached the upper acceptable carcinogenic limit (up to 1.1 × 10−4). Overall, the study highlights the importance of multi-seasonal monitoring by capturing temporally abrupt contamination events and provides a novel integrated framework that combines seasonal analysis, spatial hotspot identification, and multivariate techniques, distinguishing it from conventional single-season or non-integrated studies.
Florjana Zogaj, T. Blazhevska, F. Sallaku et al.· Limnological Review· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
