Prof Mark Lee from University of Birmingham offers panoramic view of AI’s evolution, complexities and ethical challenges

Artificial Intelligence (AI) has taken center stage across industries, driving innovation and transforming paradigms. ATLAS SkillTech University continues to pioneer AI integration not only in its B.Tech syllabus but also in design and management courses, recognising the crucial need for responsible AI implementation.

Global Discussion on Ethical AI Deployment

Recently, Prof. Mark Lee from the University of Birmingham, UK, delivered an enlightening masterclass on ‘Responsible AI’ at ATLAS SkillTech University. The session initiated a global discourse on the ethical use of AI across diverse sectors, emphasising the importance of ensuring AI’s safety and benefit to society.

Trolley Problems and Autonomous Vehicles: A Moral Conundrum

One of the captivating discussions centered around the ethical dilemmas faced by autonomous vehicles, encapsulated by the ‘trolley problem.’ Prof. Lee navigated through the complexities of decision-making scenarios for self-driving cars, presenting thought-provoking insights into responsible AI choices.

Generative AI and the Rise of ChatGPT

The masterclass delved into the realm of Generative AI, showcasing its power through deep neural networks trained on extensive datasets. The exponential growth of ChatGPT, crossing the milestone of 1 million users, exemplified the widespread adoption and influence of AI-driven conversational models.

Understanding Large Language Models (LLMs) and Bias in Text Data

Prof. Lee shed light on LLMs, elucidating their function in predicting sentence structures and revealing inherent biases in text data. The session emphasised the challenge of mitigating biases encoded in AI systems, addressing cultural norms and biases embedded within language models. One instance discussed was a university’s revelation of terminating its PhD applicant screening AI system, coinciding with its creators’ lecture on the technology. This stark scenario emphasised the pressing need to address biases within AI systems, shedding light on the complexities and ethical considerations essential in developing responsible AI solutions.

Understanding Hallucination in AI Models

Hallucination in AI models refers to instances where these systems generate content or information that lacks grounding in reality. Despite this lack of factual basis, the generated content can often be convincing, presenting a challenge in distinguishing between authentic and fabricated information. Prof Lee cited an alarming case highlighting this phenomenon. It involved a lawyer using ChatGPT during court proceedings, citing fictitious cases as part of the legal argument. This revelation has prompted the judge to consider imposing sanctions due to the misuse of AI-generated content, showcasing the potential dangers of AI hallucinations.

Intellectual Property Challenges in the AI Era

The session highlighted the complexities of intellectual property in the age of AI with the example of author Jane Friedman. Jane was astonished to discover AI-generated books listed under her name on Amazon. This unsettling instance underscores the need of robust frameworks to safeguard the rights of creators in an era where AI can replicate and create content.

Journey from Perceptrons to Transformers: Evolution in AI

From the foundational Perceptrons to the transformative impact of Transformers in Natural Language Processing (NLP), the session traced the evolution of AI models. Prof. Lee outlined the advancements in neural networks, including Multilayer Perceptrons and the resurgence of research in the 1980s and 2000s.

Engaging Q&A: Unveiling Opportunities and Challenges

The masterclass culminated in a stimulating Q&A session, where students explored AI applications, integration with MS Office and future prospects in AI-centric careers. Their inquiries highlighted the need for ongoing dialogue and education in responsible AI integration.

Prof. Mark Lee’s masterclass at ATLAS SkillTech University provided a comprehensive exploration of AI’s evolution, ethical challenges and societal impact. The university stands committed to equipping students with the knowledge and skills to navigate the responsible application of AI across various disciplines.