It's improbable that the output of any system will ever be 100% accurate; usually, AI systems are intended to represent a statistically informed prediction. These predictions, particularly in high-risk areas, may severely impact individuals and their dignity. In this session, we'll explain the statistical accuracy principle within AI systems, and host a discussion about how we should increase understanding and communicate to the public the imperfections of AI systems.
We have produced AI & Data Protection Guidance for companies that deploy AI systems. We are looking to update our guidance constantly. The audience's input and ideas will help us get a better understanding of where we need to update the guidance.
How will you deal with varying numbers of participants in your session?:Our session can flex to as many numbers as possible. We will have a host that'll help manage the conversation. With fewer numbers, we will have more of a conversation between participants; if there are greater numbers, we will use interactive tools like virtual whiteboards and surveys to engage the participants.
Ahmed Razek is an emerging technology and policy specialist. He is interested in the safe and equitable development of AI systems for the long term flourishing of humanity
Alister is the Senior Policy Officer working in the Technology department at the Information Commissioner’s Office where he focuses on AI and its implications for individual rights and freedoms.