AI Strategy and Governance
University of Pennsylvania via Coursera
In this course, you will discover AI and the strategies that are used in transforming business in order to gain a competitive advantage. You will explore the multitude of uses for AI in an enterprise setting and the tools that are available to lower the barriers to AI use. You will get a closer look at the purpose, function, and use-cases for explainable AI. This course will also provide you with the tools to build responsible AI governance algorithms as faculty dive into the large datasets that you can expect to see in an enterprise setting and how that affects the business on a greater scale. Finally, you will examine AI in the organizational structure, how AI is playing a crucial role in change management, and the risks with AI processes. By the end of this course, you will learn different strategies to recognize biases that exist within data, how to ensure that you maintain and build trust with user data and privacy, and what it takes to construct a responsible governance strategy. For additional reading, Professor Hosanagar's book "A Human’s Guide to Machine Intelligence" can be used as an additional resource for more extensive information on topics covered in this module.
- Module 1 – Economics of AI
- In this module, you will begin by examining the key inputs to AI and what tools are currently used to lower the barriers of entry for AI use. Next, you will learn the economics of AI and the competition that has emerged as AI becomes more crucial to support industry needs and we see more cloud adoption. You will learn about the value of data as it is tied to Deep Learning, and how AutoML is changing the landscape of Machine Learning, and the growing competition and implications of data harvesting. By the end of this module, you will have gained knowledge about the economic implications of AI and Machine Learning and how they impact our lives in unseen ways. You will also understand the complex nature of computational hardware and how that affects consumer demand, but also the demand for privacy.
- Module 2 – AI Innovation
- In this module, you will examine AI and data analytics to show the economical use-cases of Big Data. You will also learn about the methods and tools that are being used to lower the barriers of entry for AI use. You will review current examples of Big Data and how those firms are using their analytical tools to enhance productivity and transformation. Lastly, you will get an in-depth look at how AI can be used in BioPharma and how the payoff of their AI investment is revitalizing their industry. By the end of this module, you will have a firm grasp on the practical deployment of AI across different industries, their use-cases, and how you can best implement them to drive innovation and transformation within business.
- Module 3 – Algorithmic Bias and Fairness
- In this module, you will examine the inherent bias that can exist within data based on human behaviors. Building on these foundations, you will explore different responses within algorithmic bias and how organizations should respond and overcome these challenges. You will then review the manipulation of data, the different kinds of manipulation, and ways to ethically approach these issues. Lastly, you will examine data protection and the legal frameworks that exist to protect the consumer and individual data, and the stages of the privacy lifecycle. By the end of this module, you will have a thorough understanding of data biases, manipulation, and ethical questions of how data is handled and stored. You will be able to implement fairer algorithms and understand the legal ramifications of improperly managing data you collect.
- Module 4 – AI Governance and Explainable AI
- In this module, you will learn about explainable AI and its relationship to Deep Learning. You will also review why it is important to have explainable AI and the different approaches to creating fair algorithms and AI policies. You will also examine Explainable AI and review the necessity of equitable algorithms. You will also learn why we do not always use Explainable AI for every model, and the impacts that it can have on performance. By the end of this module, you will have gained insight into decision-making with AI and the importance of fairness and transparency in creating explainable AI systems, as well as the ethical principles and governance policies that build trust in using AI and Machine Learning.
Kartik Hosanagar, Lynn Wu, Kevin Werbach and Prasanna Tambe