How AI thinks
Learn to design smarter, fairer, and more human-centered technologies in this course which provides a comprehensive introduction to the design, development, and responsible use of modern artificial intelligence models.
With a particular focus on large language models (LLMs), students will learn the core concepts and technologies underlying these systems, including their layered architectures, in-context learning, prompt engineering, retrieval-augmented generation (RAG), and fine-tuning. Through hands-on activities, students will build a simple language model, experiment with creating functional AI tools, and critically analyze them for ethical, alignment, and explainability challenges.
Key topics include AI alignment, interpretability, and explainability, with emphasis on methods that make AI decision-making transparent and its logic accessible to users. The course also explores human oversight techniques, incorporating mechanisms such as reinforcement learning (RL) and reinforcement learning with human feedback (RLHF).
