People Shaping the Future of AI seriesVirginia Ghiara: Designing AI that empowers people
Interview article | 2026-10-30
10 minute read
AI’s greatest promise is not to replace human judgment, but to expand what people can understand, create and achieve. Making that promise a reality means confronting the ethical choices behind how AI is designed, deployed and used.
This belief sits at the heart of Research Director Virginia Ghiara’s career, connecting technology with ethics, policy and society. Her work aligns most closely with Fujitsu’s Secure & Ethical Trust in AI pillar, turning responsible AI principles into practical safeguards that help technology empower people while remaining aligned with human values.
What is Trust in AI?
Trust in AI is an approach for building confidence in AI. It is built around five key elements: Explainable, Secure & Ethical, Sovereign, Sustainable, and Connected. Together, these elements help organizations adopt AI in ways that are responsible, transparent, and aligned with human values.
In ”People Shaping the Future of AI series", we explore the personal stories, motivations, and research journeys of the people working to make trusted AI a reality. This article highlights one of the Trust in AI elements showing how one's work contributes to building AI that people, organizations, and society can trust.
From AI principles to practice
When Virginia joined Fujitsu in 2023, the responsible AI conversation was rapidly moving beyond research circles and into boardrooms, policy discussions and public debate. Organizations knew that AI should be fair, transparent and accountable, but putting those principles into practice was easier said than done.
Virginia found herself at the center of that challenge through her work with the AI4People Institute*1. It brought her together with academics, policymakers and industry experts, all approaching AI from different perspectives but with a shared desire to ensure it developed responsibly.
For Virginia, this was a chance to bring together her experience of policy, research and the real-world impact of technology to help shape conversations about the principles that should guide AI in practice. This significant work contributed to wider European debate and informed Fujitsu’s R&D activities. Together with her team, Virginia contributed to the development of Fujitsu’s AI Ethics Impact Assessment, tested solutions for mitigating bias in AI systems, and supported the development of an AI governance tool.
“We really focused on how to translate principles into practice,” she explains.
That idea became the bridge between Virginia’s earlier work and the research she leads today. As Research Director and Group Manager of The Security Science Lab at Fujitsu Research of Europe, she takes a broad view of security. Alongside traditional cybersecurity, the lab studies the risks that emerge from complex systems in which technology, people and society are closely connected, and how those risks can be anticipated and managed.
One of the initiatives Virginia is most excited to lead is “Frontria”, an international consortium bringing together researchers, industry partners and not-for-profit organizations. Together, they are researching some of the most pressing challenges surrounding AI-enabled disinformation, AI governance, trust and security. Frontria reflects the collaborative approach that has defined Virginia’s work. From her early involvement with AI4People to the international partnerships she leads today, Virginia has consistently brought different perspectives together to turn difficult questions about technology and human values into practical action.
*1: The AI4People Institute brings together academia, global businesses, civil society organizations and governments to consider the risks of AI and advocate for its responsible and beneficial development and deployment. Through its Ethical Framework for a Good AI Society, AI4People helped initiate the regulatory process that led to the AI Act in Europe, the world’s first comprehensive AI regulation.
An unconventional path to AI
Virginia’s path into AI did not begin with coding or computer science. It began with a habit of asking why.
As a child growing up in Italy, she was curious about how people made decisions, what influenced them and how culture shaped the way people understood the world. Those questions led her to philosophy and, eventually, to the study of knowledge, evidence and ethics.
She studied at the University of Turin before pursuing interdisciplinary studies spanning government, science, philosophy and economics. Her education subsequently took her to Rotterdam and then to the University of Kent for a PhD, followed by a research period in the School of Public Health at Imperial College London.
Her path may have been unconventional, but each move gave her a new perspective.
“I always wanted to use philosophy to have an impact in the world, not just to have theoretical discussions.”
After her academic research, she moved into evidence-based public policy, where her work included supporting decisions connected with the UK Department for Work and Pensions. As AI became increasingly relevant to areas such as children’s social care, the questions she had explored through philosophy began to take on an immediate practical importance.
Despite moving between philosophy, economics, science, policy and industry, Virginia sees one continuous thread.
“AI is not just something technical. It is really linked to fundamental questions related to responsibility, human values and what kind of society we want to build.”
There’s no “one-size-fits-all” approach to trust
Virginia’s understanding of AI has also been shaped by working and studying across Italy, the Netherlands, the United Kingdom and the United States. These experiences showed her that trust in AI is influenced not only by technology itself, but also by cultural attitudes, public policy, and societies' previous encounters with AI systems. In her experience, AI use appeared more widespread in the UK than in Italy. At the same time, conversations she encountered in Italy often focused more strongly on the ethical and societal implications of AI.
While anecdotal, these contrasts reinforced her appreciation of how cultural and policy contexts shape public trust in technology.
“Trust is not a single characteristic that can be measured in the same way across all AI systems.”
In Virginia's view, the qualities that make a system trustworthy depend on how it is used, who is affected by it, and what risks it creates. For an AI system protecting critical infrastructure, trust may depend primarily on security and resilience. For a system supporting hiring or access to public services, fairness and the absence of bias may be paramount. In other contexts, explainability may matter most.
It is therefore important to understand the expectations of users, customers, regulators, and the people affected by the system, and build the appropriate safeguards from the outset. This is why Virginia advocates a "trust by design" approach. Rather than applying a generic framework, organizations should identify what trust requires in a specific context and embed those requirements throughout the development process. As she explains, “You cannot treat trust as a checklist.”
Turning research into real-world impact
One achievement that stands out for Virginia is helping Fujitsu’s responsible AI research move beyond the lab and into a platform that organizations can use. It is a real-world example of “trust by design.”
Fujitsu licensed five AI Trust technologies from its Fujitsu Kozuchi AI service to Italian start-up AKOS AI. They now underpin AKOS HUB, a no-code and low-code governance platform that helps enterprises classify AI risk, meet EU AI Act requirements, assess ethical impacts, and identify and mitigate bias across the AI lifecycle.
“We managed to go to market with our technologies, and now they can be used by other people,” Virginia says. For her, that is the real significance: principles such as fairness, accountability and responsible governance becoming tools that can guide how AI is developed and deployed.
Empowering people in an AI-enabled world
Virginia first encountered AI through her work in philosophy and data-driven science more than a decade ago. What surprises her now is not only how far the technology has advanced, but also how readily people have integrated it into their daily lives, often placing their trust in systems they do not fully understand.
“What I didn’t anticipate at the time was the spread of AI and how fast it would have changed the way in which we live.”
But as AI’s presence grows, so does the need to use it responsibly. AI systems can produce convincing but inaccurate information, introduce security weaknesses, and encourage overreliance on automated outputs. For Virginia, one of the greatest risks is anthropomorphizing these systems: when people begin to treat chatbots as if they think, feel, or understand the world as humans do, they may grant them a level of trust that is neither justified nor deserved.
For Virginia, the answer is not to reject AI, but to be clearer about the role we want it to play: using it to extend human capabilities while keeping people responsible for judgment and decisions.
“How can we be empowered by AI rather than how can we replace our work with AI?”
For Virginia, empowering people is not only about supporting individual tasks. AI can give researchers, organizations and societies new ways to understand complex challenges and respond to them. In the Security Science Lab, for instance, the team is developing an Ocean Digital Twin: a digital representation of the ocean that could help researchers analyze environmental change and inform responses to climate change. She sees similar potential in scientific discovery, healthcare screening and tackling disinformation.
Defining the future of trustworthy AI
Fujitsu defines Trust in AI through five interconnected elements: Explainable, Secure & Ethical, Sovereign, Connected and Sustainable. Virginia’s work sits most directly within Secure & Ethical, where responsible AI, security and disinformation come together.
“Something which is quite unique in Fujitsu’s approach is that trust in AI is treated as a goal in itself. It’s not a sort of by-product.”
That principle has guided Virginia from philosophy and policy into research and commercialization. It also shapes what comes next. Her priorities include misinformation and disinformation, challenges that AI can both amplify and help mitigate, and the security and safety of emerging AI systems. While she recognizes the risks associated with AI, she is cautious about framing the technology as an existential threat. Instead, she argues for practical safeguards, standards, and appropriate regulation to guide its development and use.
For Virginia, AI’s future depends on designing technology that responds to real human needs and strengthens rather than replaces human judgment. It is an ambition she captures simply:
“To develop AI that can empower people.”
Related Information
From questioning an unjust world to building ethical AI
