LinkedIn is using responsible AI to create economic opportunities for its members across the world. The company focuses on best practices in fairness and privacy to build a better and more equitable user experience.
- LinkedIn is applying responsible AI to create economic opportunities for its members across 200 countries worldwide
- The company focuses on best practices in fairness and privacy to build a better and more equitable user experience
- LinkedIn follows Microsoft's responsible AI principles and the six values that they strive to build into their products
- LinkedIn has made the most progress on fairness and privacy
- Fairness is a hard problem to solve, but LinkedIn has developed a framework to quantify it
- LinkedIn uses experimentation to ensure that changes made to their product benefit every member and do not introduce unintended consequences
Ya Xu, VP of Engineering and Head of Data Science at LinkedIn, shared how the company used experimentation to evaluate whether they were introducing any unintended consequences when making changes to their product. They looked at treatment effects by categories and other important attributes such as network size to ensure that they were not shifting the distribution of opportunities. This approach allowed LinkedIn to introduce fairness awareness in every change they made to their product.