At a time when artificial intelligence is becoming part of everyday life, conversations about trust, safety, and responsibility are becoming impossible to ignore. Those themes took center stage at the International Conference on Data Science and AI for Social Good and Responsible Innovation (DASGRI 2026), hosted by the School of Computing at Goldsmiths, University of London.
Among the invited speakers was Arun Kumar Elengovan, a cybersecurity leader whose work spans applied cryptography, cloud identity systems and enterprise scale security engineering. With more than fifteen years of experience across distributed systems and digital trust infrastructure, Elengovan has contributed to initiatives focused on cloud security resilience, identity protection, and secure system design.
He is also an IEEE Senior Member, a Fellow of IETE, and a Fellow of BCS and regularly participates in international conferences, peer review activities and professional technology forums. Rather than focusing only on technical complexity, his session explored something more fundamental: what it takes for people to trust artificial intelligence systems that increasingly influence modern society.
Speaking to researchers, students, and technology professionals attending the conference, Elengovan described trust as the foundation upon which every digital system ultimately depends. As artificial intelligence expands into healthcare, financial systems, public infrastructure, and identity platforms, he explained that security failures are no longer isolated technical problems. In many cases, they carry real human consequences.
“Responsible innovation cannot exist without security,” he told attendees. “When systems that people depend on are compromised, the damage goes far beyond technology.”
Throughout the session, he emphasised that many modern threats begin quietly. Instead of dramatic attacks, failures often emerge through overlooked weaknesses, manipulated datasets, insufficient visibility or systems that were never designed to adapt under pressure.
One area that drew particular attention was the growing risk of data manipulation in artificial intelligence models. Elengovan explained how compromised training data can gradually influence the behavior of AI systems, creating unreliable or biased outcomes without immediate visibility. In environments such as healthcare or public services, he noted, even subtle failures can have wide reaching societal impact.
At the same time, the discussion remained practical rather than alarmist. Elengovan spoke about the importance of building resilient systems through stronger data integrity practices, adversarial testing, privacy preserving technologies and continuous monitoring. He encouraged organisations to treat cybersecurity and ethical AI not as separate conversations, but as closely connected responsibilities.
The session also reflected on the importance of collaboration between academia and industry. According to Elengovan, universities play a critical role in shaping the future of responsible technology, while industry practitioners provide real world perspectives on how modern systems behave under operational stress.
“Technology evolves very quickly,” he shared during the discussion, “but trust takes much longer to build and only moments to lose.”
For many attending DASGRI 2026, the conversation served as a reminder that the future of artificial intelligence will not be defined solely by capability or speed. It will also depend on whether societies can design systems that remain secure, resilient, and worthy of public trust.
