Privacy by Design and Default in AI: The Role of Privacy-enhancing Technologies

As AI continues to evolve, privacy concerns surrounding the collection and processing of personal data in AI systems, especially in models like generative AI, are growing. Privacy-Enhancing Technologies (PETs) hold immense potential to help implement privacy by design and by default when developing AI systems and enable broader beneficial sharing and use of data across different companies and sectors to boost further AI adoption. In this panel, experts will explore how PETs, such as federated learning, differential privacy and homomorphic encryption, can both strengthen privacy protection and drive innovation in AI. CIPL will also share insights from its paper, “PETs and PPTs in AI: Enabling Data Use and Operationalizing Privacy by Design and Default,” outlining how these technologies address privacy risks and create new opportunities in data sourcing, model training, security and collaboration.

What you will learn:

• How PETs such as differential privacy, federated learning and homomorphic encryption can enhance privacy and foster innovation, through real-world industry use cases.
• The key challenges companies face in implementing these technologies and strategies for overcoming them, including insights into the evolving regulatory and ethical landscape. 
• What actionable steps can be taken to encourage widespread adoption of PETs and advance a more secure, trustworthy AI ecosystem.