Foundations of Responsible AI (2022)

Foundations of Responsible AI (2022)

2h 31mIntermediate2022-08-02

Authors

Ayodele Odubela

Ayodele Odubela

Data Scientist and AI Ethicist

Course details

How well do you understand the issues around algorithmic bias and unfairness? In this course, data scientist and AI ethicist Ayodele Odubela teaches you the principles of responsible AI, as well as the frameworks necessary to apply RAI techniques in AI systems. Ayodele explains modern AI development and problems in machine learning (ML) that differ from software engineering. She discusses big data and where it comes from, then covers several important points of data awareness and literacy. Ayodele goes over ethical frameworks, consequence scanning, fairness, accountability, and more. She also explains fairness-related harms, why they arise, why they matter, and what you can to avoid or mitigate them. Plus, Ayodele offers a detailed description of human rights as related to AI.

Skills covered

Responsible AIArtificial Intelligence FoundationsFoundationsArtificial Intelligence (AI)

Concepts

Introduction

  • Understanding responsible AI

Philosophy of AI

  • What is AI and how does data enable it
  • Modern AI development
  • Problems in ML that differ from software engineering

Data Awareness and Literacy

  • Big data and where it comes from
  • Seeing trends in data
  • Building data understanding
  • Visualization and comparing data
  • Storytelling with data

Ethical Theories

  • Introduction to ethical AI
  • Ethical frameworks
  • Beneficence vs. maleficence
  • Calculating consequences
  • Consequence scanning
  • Common good and equity

Responsible AI Principles

  • Fairness
  • Transparency
  • Accountability
  • Explanations
  • Interpretability
  • Inclusivity

Algorithmic Harm

  • Why fairness related harms
  • Critical AI incidents and learnings
  • Bias in the design and development lifecycle
  • Causal reasoning and fairness
  • Risk mitigation in AI
  • Technical aspects of sociotechnical solutions

Human Rights and AI

  • Anonymity and data privacy
  • Unintended uses and misuses
  • Unethical business cases
  • Autonomous systems and society
  • Who AI is developed for

Conclusion

  • AI regulation and applying responsible AI frameworks
80,000 Toman