5 questions to...
Vito Morreale
Interview with the Senior Technical Manager AI & Data R&I of Engineering.
Vito Morreale leads the AI & Data Research & Innovation Lab at the Engineering Group. The Lab's mission is to advance the state of the art in the development of innovative systems based on artificial intelligence, big data, and advanced analytics, with applications in critical sectors such as industry, physical and cyber security, defense, agri-food, and infrastructure.
The AI & Data R&I Lab is structured into several technological and application-focused research areas. It includes around 130 researchers and is involved in more than 60 national and European research projects, collaborating with hundreds of companies, universities, research centers, and public institutions.
A Senior Manager with over 20 years of experience leading Research, Development, and Innovation teams and laboratories, Vito is an active member of several boards and working groups dedicated to research and innovation.
1. WHAT ARE THE MAIN RESEARCH AREAS ENG IS EXPLORING IN ARTIFICIAL INTELLIGENCE AND ADVANCED DATA ANALYTICS?
The AI & Data Lab at Engineering is at the forefront of data and artificial intelligence technologies. By combining medium-term industrial research with a long-term strategic vision, our goal is to support delivery teams with advanced skills and tools to develop increasingly innovative, state-of-the-art solutions for our clients — while also optimizing our internal processes.
From a technological perspective, our research areas are grouped into three main domains: Data Management, AI, and Decision Intelligence.
In the Data Management area, we work on core topics such as data governance and architecture, modeling, harmonization, and integration from heterogeneous sources. We are studying and applying emerging concepts and models related to data spaces and the data economy, with a focus on data quality, metadata, and semantics. One key area is the integration and enhancement of data from edge and IoT sources. Additionally, we are developing solutions for the discovery and integration of data sources from the Internet, including the deeper layers of the Dark Web and DarkNet, with a strong emphasis on digital asset chain of custody.
In the Artificial Intelligence domain, we explore a broad range of technologies, techniques, and methodologies, adopting an approach that blends science with engineering. We place strong emphasis on Generative AI based on Large Language Models (and soon, Large Action Models) and Agentic AI. We pay particular attention to model-centric AI, which focuses on improving machine learning models rather than solely on data. In this context, we are also developing methods and tools for Federated Machine Learning (FML) — a distributed training paradigm where data remains locally within the systems that generate and manage it, ensuring privacy, security, efficiency, and data sovereignty. We also work on data-centric AI, which emphasizes data quality, consistency, and representativeness over continuous model enhancement. Furthermore, we are investing in composite AI, a paradigm that combines various AI techniques and approaches to overcome individual limitations and build more effective, robust, and adaptive intelligent systems.
Finally, we develop advanced solutions in Decision Intelligence, an interdisciplinary approach that integrates data, models, AI, and decision sciences to support and enhance both human and automated decision-making. In this area, we also work on data storytelling, interactive dashboards, visual analytics, and geospatial intelligence, all aimed at empowering decision support systems within the framework of augmented intelligence.
2. WHAT ARE THE CHALLENGES IN TRANSLATING AI & DATA LAB RESEARCH PROJECT RESULTS INTO MINIMUM VIABLE PRODUCTS (MVPs) THAT MEET BUSINESS NEEDS?
Our research is applied and impact-oriented, aiming to create real value across key sectors for Italy and Europe, where our technological solutions are adopted and implemented in real or highly realistic use cases and scenarios. As the AI & Data R&I Lab, our role is to act as a bridge between vision and execution, between scientific potential and tangible value. There are several challenges we face when driving AI & Data research toward business application.
One of the first hurdles is correctly interpreting business needs, which are not always clearly defined or expressed in terms of innovation. Over the years, our approach has allowed us to build both technological and domain-specific expertise, understand process dynamics, and steer research toward the development of assets (know-how, PoCs, prototypes, use cases). This is made possible through constant dialogue with multiple stakeholders to translate real-world needs and emerging technologies into data- and AI-based solutions.
Another key challenge is the mismatch between research timelines and business expectations. Research typically looks to the mid-to-long term, while business operates on much shorter decision-making and execution cycles, demanding faster results. That’s why it is essential to design innovation programs in AI & Data - even in specific market sectors - that define progressive milestones (PoCs, prototypes, MVPs) within a long-term vision grounded in concrete outcomes.
Research results and early-stage solutions must also be conceived from the outset with adequate maturity and technical sustainability in mind. Innovating with AI also means ensuring solutions are reliable, scalable, and secure. Building an effective MVP requires adopting an engineering mindset from the prototype stage, using modular architectures, real-world data testing, and post-deployment monitoring. This is particularly challenging today, as AI engineering practices are still maturing across the industry.
Finally, in the field of AI and data management, compliance with regulations, policies, and both general and sector-specific standards (e.g. GDPR, AI Act) is non-negotiable. Issues like privacy, algorithmic transparency, non-discrimination, accountability, and decision traceability must be addressed from the very early stages. Our team includes professionals with legal and ethical compliance expertise and collaborates with external partners to build explainable and compliant assets. We also work with European institutions on topics related to ethics and responsibility in the use of AI and data management - including in sensitive domains like healthcare and public safety.
3. WHICH OF YOUR LAB'S CURRENT RESEARCH PROJECTS ARE HAVING THE GREATEST IMPACT, AND IN WHICH APPLICATION DOMAINS?
Currently, the projects within the AI & Data research lab address a wide range of technological and application-oriented topics across various sectors.
In the industrial sector, we are working on projects that support the transition to Industry 5.0. The CircularTwAIn project focuses on circularity in production chains and the lifecycle of industrial products, using AI-powered Digital Twins. We have developed and are continuously evolving the TRUE Connector, our open-source solution for secure and reliable data exchange, aligned with the IDS standard.
In the agri-food sector, the AgriBIT project integrates high-resolution geospatial data, drone imagery, and IoT sensor data with AI techniques to reduce costs and environmental impact, while increasing crop yield.
In healthcare, we’re developing solutions for prevention, early diagnosis, personalized treatments, and system efficiency, while also addressing ethical and regulatory challenges. For example, in the SHE project, we explore precision medicine by integrating omics data with electronic health records.
In the cybersecurity domain, the CyberSEAS project has focused on protecting electric power infrastructures from high-impact cyberattacks. It led to advancements in our RATING cyber risk assessment platform and enhanced tools to counter social engineering threats. We're also exploring IoT and AI system security in the ERATOSTHENES, CERTIFY, and KINAITICS projects.
In the area of physical security, projects like ARIEN (targeting drug trafficking) and others - including AIDA, STARLIGHT, ANITA, and GRACE - address the fight against crimes such as cybercrime, terrorism, sexual exploitation, and illegal trafficking. Physical and cyber security intersect in the protection and resilience of critical infrastructures. The ATLANTIS project focuses on natural hazards and complex cyber-physical-human threats across various sectors and European countries.
We are also developing cross-cutting solutions. The HAIR project* involves building, training, and integrating a full suite of multilingual Large Language Models (LLMs) as part of our secure and private Generative AI platform, EngGPT.
Finally, in the strategic AVANT project (part of the IPCEI-CIS program), we are developing and applying data and AI technologies within a multi-platform framework for building Digital Twin applications and systems across all vertical sectors.
*The HAIR project is funded by Italy’s National Recovery and Resilience Plan (PNRR), Spoke 0 – Supercomputing Cloud Infrastructure.
4. ENG'S COMMITMENT TO AI RESEARCH HAS BEEN WELL-ESTABLISHED FOR YEARS. WHAT HAVE BEEN THE KEY MILESTONES ON THIS JOURNEY SO FAR?
Since the early 2000s, ENG has focused its research on Cloud Computing, Security, Knowledge Management, and Intelligent Systems. It was during this time that we began to consolidate our dedicated team for studying autonomous systems. As early as 2002, we were working on multi-agent projects in areas such as e-commerce, tourism, and finance. In 2005, we developed the PRACTIONIST framework to tackle complex problems using collaborative agents.
Over the following two decades, our participation in European research programs intensified, culminating in the last ten years with outstanding results that have positioned ENGINEERING among the leading players in innovation both in Italy and across Europe.
We then turned our focus toward the processing and enhancement of big data, which had become a key enabler for intelligent solutions across nearly every industry. Deep learning, with its ability to learn hierarchical data representations, led to major breakthroughs in fields such as natural language processing and large-scale prediction. In parallel, graph intelligence emerged strongly - using AI to analyze complex data structures such as networks of relationships, knowledge, or events. This approach supports a wide range of applications, from recommendation systems and crime detection and investigation, to bioinformatics and the analysis of social or industrial phenomena.
Two major trends have redefined the landscape of our AI research in recent years: the rise of Generative AI and the Agentic AI paradigm. We are exploring new methodologies and techniques to build innovative, sophisticated Large Language Models (LLMs) that are natively compliant with evolving regulations, while remaining efficient and sustainable. At the same time, we contribute to the development of AI systems based on autonomous agents, aiming to reach a level of maturity and reliability comparable to that of conventional software.
Lastly, we are closely observing how the distribution and/or federation of both data and computation can enable a new generation of intelligent systems - ones that are proactive, autonomous, and capable of making decisions in complex environments (multi-actor, multi-platform, multi-technology, etc.).
Throughout this journey, we have completed hundreds of projects, collaborating with numerous universities, research centers, companies, and public institutions in Italy and internationally. Through these efforts, we have cultivated and promoted a strong culture of technological and applied innovation - always driven by the spirit and goal of “staying one step ahead and ready for what’s next.”
5. WHICH ASPECTS OF SCIENTIFIC RESEARCH WILL BE PRIORITIES IN THE COMING YEARS TO SUPPORT THE EVOLUTION OF AI AND INTELLIGENT SYSTEMS?
AI is redefining internal operations, business models, and competitive dynamics, pushing companies toward an AI-first approach with scalable solutions. It’s no longer just about using AI, but about designing it from the outset as a core pillar of business strategy.
In this context, the future holds numerous and complex challenges. Among the most critical is AI governance, essential for ensuring transparency, compliance, and trust. To manage bias and the risk of privacy violations, we’ll need robust Responsible AI frameworks, supported by ethical oversight. As Agentic AI evolves, the demand for reliability, robustness, and security becomes even greater, especially in high-stakes environments. Intelligent systems must be resilient to errors and attacks, and able to operate in dynamic, unpredictable contexts.
Many of these challenges converge into what is now emerging as AI Engineering - a growing discipline that encompasses the practices, methods, and tools needed to design, develop, deploy, and maintain AI systems that are robust, scalable, reliable, and sustainable. AI Engineering brings together software engineering, data science, machine learning, MLOps, DevOps, data engineering, system architectures, data governance, ethics, and compliance — enabling the creation of AI solutions that are reproducible, maintainable, monitorable, and updatable.
Within this framework, our research and innovation efforts - aligned with our corporate AI strategy - are focused on building skills and tools across three core dimensions:
- Our proprietary LLM suite, EngGPT, which forms the beating heart of our AI capabilities.
- AI Engineering, meaning the development of a full ecosystem of architectures, tools, and frameworks - built around our ALIDA platform - to design, deploy, and monitor AI solutions that are technically sound, scalable, and ready for industrial adoption.
- Generative AI tools for software engineering, designed to evolve the application development lifecycle by significantly increasing productivity and reducing time-to-market - not only through automation, but by enabling true augmented intelligence for developers.
These three pillars - proprietary LLMs, AI Engineering, and GenAI for software development - form the foundation of our vision: deeply interconnected, mutually reinforcing, and fully dedicated to enabling the digital transformation of our clients.
We have cultivated and promoted a strong culture of technological and applied innovation - always driven by the spirit and goal of “staying one step ahead and ready for what’s next".
Vito Morreale
Senior Technical Manager AI & Data R&I of Engineering
Recommended for you
Explore additional content associated with the topic
