Building trust with community-informed AI evaluation


Summary

The video explores the critical role of building trust for adopting generative AI and enhancing user experiences, with a focus on engaging marginalized communities in the development process. It delves into social technical approaches and the importance of AI principles such as fairness, transparency, and accountability in guiding responsible AI development. The discussion emphasizes community-informed methods to grasp global complexities, respect local nuances, and move beyond generic solutions in AI research, advocating for equity-based approaches to address systemic disparities and empower affected communities. The video also provides a step-by-step guide on creating adversarial queries, involving expert feedback, identifying sensitive domains, equity concerns, and model risks in AI evaluation.


Introduction to Building Trust with Community-Informed AI Evaluation

Discussion on the importance of building trust for generative AI adoption and better user experience, focusing on engaging with marginalized communities to ensure responsible development of AI tools.

Social Technical Approaches in AI Research

Overview of social technical approaches in AI research and the development of AI principles like fairness, transparency, and accountability to guide responsible AI development.

Community-Informed Methods in AI Research

Explanation of community-informed methods as a key approach to understanding global complexities, respecting local nuances, and moving beyond generic answers in AI research.

Defining Community for Community-Informed Practices

Discussion on defining community beyond geography and understanding communities based on shared concerns and passions to establish partnerships and collaborations in research.

Equity-Based Approaches in AI Research

Exploration of equity-based approaches in AI research, emphasizing understanding systemic factors contributing to disparities, data-driven decisions, and empowering affected communities.

Creating Adversarial Queries for AI Evaluation

Step-by-step guide on creating adversarial queries by incorporating expert feedback, identifying sensitive domains, key subdomains, equity concerns, sensitive user groups, use cases, and model risks in AI evaluation.

Logo

Get your own AI Agent Today

Thousands of businesses worldwide are using Chaindesk Generative AI platform.
Don't get left behind - start building your own custom AI chatbot now!