The DAWL AI Lab is a research laboratory within the Department of Artificial Intelligence at the University of Malta. Led by Dr Dylan Seychell, the lab is based in FICT Lab 0A23 and serves as a collaborative space where researchers, students, and practitioners work together to explore, develop, and apply Artificial Intelligence for real-world impact.
The lab takes its name from the Maltese word "Dawl", meaning "light". The name reflects several dimensions of our mission: the pursuit and dissemination of knowledge, the promotion of AI and media literacy through informed public engagement, and the central role of light itself in Computer Vision and imaging systems. As both a symbol and a scientific principle, dawl represents discovery, understanding, and the ability to perceive and interpret the world through intelligent technologies.
Our work spans Computer Vision, Natural Language Processing, Geospatial Intelligence, Robotics, and Multimodal AI, with a focus on addressing practical challenges that matter to society. Through research, innovation, and community engagement, we aim to advance the responsible development and adoption of Artificial Intelligence in Malta and beyond.
Research and development activities are organised around three complementary streams: AI for Environment, AI for Media Analysis, and AI Literacy. Together, these streams combine methodological innovation with publicly accessible tools, datasets, educational resources, and real-world applications that support both scientific progress and public understanding of AI.
Whether you are a researcher, student, industry collaborator, policymaker, or simply interested in the future of AI, we welcome you to explore our projects, publications, and initiatives.
Click any research area to explore projects, publications, and initiatives
🔗 Website: https://aienvironment.org/
Learn more about this research stream
The AI for Environment stream develops AI systems for environmental monitoring, urban management, geospatial intelligence, and environmental decision support. Research spans satellite and aerial imagery, UAV-based sensing, vehicle-mounted perception systems, and ground-level Computer Vision, enabling the observation, analysis, and management of environmental and infrastructure-related challenges across multiple spatial and temporal scales.
EMBAT · Enhancing Malta's Basemap with AI Technology
AI-assisted geospatial analysis for national basemap maintenance, including building footprint extraction, temporal change detection, shadow-aware processing, and GIS-ready decision support derived from aerial and satellite imagery.
AWIGS · Aerial Waste Identification and Geolocation System
Drone-based detection, classification, and geolocation of litter from aerial imagery, supporting environmental monitoring and waste-management operations in difficult-to-access locations.
TALC · Targeted Aerial Litter Collection
An extension of AWIGS investigating autonomous litter retrieval through the integration of UAVs, adaptive robotic grippers, and AI-based decision-making for safe and effective collection.
AAC · Automated Aerial Capture for Enhanced 3D Data Acquisition
AI-driven adaptive flight planning and image acquisition for high-quality 3D reconstruction, digital-twin generation, and large-scale environmental modelling.
AICOM · Application of AI and Computer Vision to Optimise Cleansing Operations
Privacy-aware computer vision systems for urban cleansing and maintenance operations, supporting waste detection, infrastructure monitoring, operational planning, and evidence-based decision-making.
Environmental challenges rarely exist within a single perspective or sensing modality. By combining AI and Computer Vision capabilities across satellite, drone, and ground-based platforms, this research stream develops integrated environmental intelligence systems capable of providing richer insights, improved situational awareness, and more effective support for environmental management, infrastructure maintenance, and public-sector decision-making.
🔗 Website: https://ainewsanalysis.org/
Learn more about this research stream
The AI for Media Analysis stream develops AI systems that support media transparency, media literacy, representation analysis, bias detection, and data-informed journalism. Research combines Natural Language Processing, Computer Vision, Multimodal Video Analysis, Information Retrieval, and Explainable AI to better understand how information is produced, presented, and consumed across contemporary media ecosystems.
By developing tools that help analyse media content at scale, this stream supports researchers, journalists, policymakers, regulators, educators, and citizens in making sense of increasingly complex information environments while promoting transparency, accountability, and responsible AI adoption.
DIMAS · Data-Informed Media Analysis Suite
An AI-powered platform for large-scale analysis of Maltese media content, including sentiment analysis, named entity recognition, semantic search, topic exploration, bias insight generation, and role-specific tools for journalists, researchers, and the public.
AIRAS · AI-Driven Media Representation Analysis for Social Equity
AI-based analysis of gender representation and sentiment in Maltese news, supporting article-level assessment and dashboard-based exploration of representation patterns across outlets, topics, and time.
TMMU · Transparent Media Monitoring and Understanding
Multimodal analysis of Maltese news videos using visual, textual, and audio information to support media transparency, representational analysis, structural pattern recognition, and regulatory oversight.
NBxAI · Exploring Visual Bias in News Content using Explainable AI
Investigation of potential visual bias in news reporting through the analysis of article text, captions, headlines, generated image descriptions, and image-text relationships using computer vision, Natural Language Processing, and explainable AI techniques.
Modern media exists across text, images, video, and audio. By combining advances in Natural Language Processing, Computer Vision, Multimodal Learning, and Explainable AI, this research stream develops tools and methodologies that help make media ecosystems more transparent, measurable, and accountable. Through collaboration with journalists, researchers, public institutions, and civil society, the stream seeks to strengthen media literacy, support evidence-based decision-making, and foster greater public understanding of how information is created, distributed, and interpreted.
🔗 Website: https://ai4all.mt/
Learn more about this research stream
The AI Literacy stream focuses on advancing public understanding, responsible use, and informed adoption of Artificial Intelligence. Through educational initiatives, community engagement, curriculum development, and applied research, this stream seeks to empower individuals and organisations to navigate an increasingly AI-enabled world with confidence, critical awareness, and practical skills.
Recognising that the societal impact of AI extends beyond technical development, this stream explores how AI knowledge can be made accessible to diverse audiences while promoting ethical, safe, and responsible use of emerging technologies.
Project NAIL · National AI Literacy Initiative
Project NAIL (National AI Literacy) is a nationwide educational initiative designed to help the Maltese public understand and use Artificial Intelligence responsibly. Developed by the Malta Digital Innovation Authority (MDIA) in collaboration with the University of Malta and industry partners, the programme provides free and accessible AI training for residents through the AI for All (AI għal Kulħadd) platform.
The programme requires no prior knowledge of Artificial Intelligence and is delivered through an online learning environment accessible using a Maltese eID. Learning pathways are designed to support diverse audiences, including students, educators, job seekers, entrepreneurs, older adults, and individuals interested in independent living.
Key learning areas include:
- AI Fundamentals and Critical Use — understanding how AI systems work, recognising limitations and unreliable outputs, and protecting privacy and personal data.
- AI for Everyday Life — practical applications of AI in daily personal and professional activities.
- AI for Learning — responsible use of AI as a tool for education, research, and lifelong learning.
- Audience-Specific Learning Pathways — tailored modules addressing the needs of different demographic and professional groups.
Participants receive digital certification upon successful completion of programme modules and gain access to AI tools that support continued learning and practical experimentation.
AI literacy is a critical component of responsible AI adoption. By combining educational resources, public engagement activities, research, and practical training opportunities, this stream aims to strengthen digital resilience, foster informed participation in AI-enabled societies, and ensure that the benefits of Artificial Intelligence remain accessible to all.
This GitHub organisation hosts outputs from ongoing and completed applied AI projects. The availability of source code, models, datasets, and documentation may vary between repositories due to publication schedules, research collaborations, data protection requirements, licensing constraints, or operational considerations.
We are committed to supporting transparency, reproducibility, and knowledge sharing wherever possible, and additional resources may be released as projects mature.
University of Malta · DAWL AI Research Lab · Department of AI
