Introduction to AI Ethics and Safety

There’s a lot of hype around AI, both positive and negative. Some experts think AI will bring us into a new utopian age, and others think it will bring about humanity’s destruction. This course aims to explore the potential biases and hidden consequences of AI models, as well as address questions and concerns regarding AI risks. Instructor Kate Harwood covers case studies, provides insights on manipulating AI output, and teaches you how to identify biased datasets and models. Additionally, learn how to think critically about the development of large AI models and discuss thought-provoking questions related to AI's understanding of the world and of sentience. Learn what can make an AI model biased, discover how models can have hidden consequences, and explore the nuances of AI risks, from the practical to the profound.

Learning objectives
Spot biased datasets and models in AI systems, ensuring fairness and accuracy.
Manipulate AI outputs using methods to recognize and mitigate risks of misinformation or bias in AI-generated content.
Navigate the complexities of aligning AI systems with human values and goals, exploring ethical and technical considerations.