LLMs are driving the frontiers in computer performance today. This talk explores the MLPerf LLM benchmark landscape, the unique challenges of building LLM benchmarks for training and inference, and the challenges for submitters.

Billions of people around the world rely on access to information through websites and mobile applications. We are building for a privacy-first future, one where people have transparency in their ability to keep their online information private, and at the same time ensuring we maintain the integrity of a free and open internet. This requires a very close collaboration between Governments, Industries and Digital Ecosystems.

The IMDA x Google: PETs x Privacy Sandbox initiative; we are helping companies migrate to this privacy-first world, by providing the education and tooling necessary to become privacy-ready. In this session we will provide the audience with an overview of this strategic programme, discuss key updates, and share recommended actions to help prepare for a privacy-first future. 

Digital transformation is essential for government agencies to stay competitive and deliver better services to citizens. But with the increasing amount of data being collected, proper attention to data privacy concerns is crucial to maintain the trust of stakeholders. This talk will explore the challenges and opportunities for promulgating PETs in the public sector, and the role of PETs can fit into organisations’ data engineering strategy. In particular, it will touch on some of the lessons learned in developing a Central Privacy Toolkit, as part of a wider ecosystem of data engineering tools that allow public sector officers to collaborate, share and exploit data while striking the right balance between data-driven innovation and privacy protection.