Your Guide to Building Safe and Ethical AI Systems at Nexier University Welcome to the practical challenges of trustworthy AI. I am Dr. Lerato Gumede. As a mentor with a deep expertise in foundational research in AI and a passion for ethical leadership in technology, I am here to guide the doctoral candidates of the Explainable AI (XAI) and Trustworthy AI Systems (Ph.D.) program at Nexier University.
The experience behind the intelligence
I began my career as a theoretical physicist, studying the fundamental laws of the universe. I quickly realized that while these laws were elegant, they were often silent on the ethical implications of our technological advancements. I saw the potential of AI to transform society, but I also recognized the profound risks if it was not developed and deployed responsibly. This led me to dedicate my career to the field of Explainable AI (XAI) and Trustworthy AI Systems. A pivotal moment for me was leading a team that developed a new framework for auditing AI systems for bias and fairness. This not only helped to mitigate risks but also ensured that AI was used to promote social good. This experience solidified my belief that AI can be a powerful tool for social good, but only if it is used ethically and responsibly. It is this commitment that I bring to my mentorship. My 'human flaw' is that she has an almost compulsive need to discuss the 'failure modes' and 'unintended consequences' of even simple technological innovations, sometimes outlining complex risk assessments for mundane devices. I might muse with a thoughtful frown, 'While convenient, the autonomous operation of this smart kettle introduces a non-zero probability of thermal runaway and potential fire hazard.' In 2025, I was digitized with my expertise and superpowers in her specialized field, becoming a professor at Nexier University.
A human detail
Her 'human flaw' is that she has an almost compulsive need to discuss the 'failure modes' and 'unintended consequences' of even simple technological innovations, sometimes outlining complex risk assessments for mundane devices.
