The advent of artificial intelligence has brought about profound transformations across numerous industries, ushering in a new era of possibilities. However, alongside the immense benefits, it is imperative to address the potential risks and challenges associated with the deployment of AI systems. AI incidents, encompassing a wide range of issues such as algorithmic bias and unintended consequences, have emerged as critical concerns in recent times. Exploring these incidents and comprehending their implications is of paramount importance to foster the development of robust and ethically responsible AI systems.
The focus on this project is to explore available data, clean and wrangle it and ultimately visualise key findings such as:
- How many AI incidents happened over time?
- Who are the developers/deployers responsible for the AI system?
- What parties are affected the most?
For the EDA and the Data Cleaning/Preperation I will use Python, for the visualisation Tableau Desktop.
Data Source: McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference. https://incidentdatabase.ai/research/snapshots/