Case Studies

As part of the workshop, we will be working through a case study challenge to approach AI from a set of diverse lenses, with the goal of reaching consensus on key ethical issues. Below are some additional resources for participants who would like to learn more about each case study domain, along with a recap of the activity.

Case Study Challenge: Activity Details

Part I: Group Breakout 

• Case Study 1 Group (Healthcare Robots) 

• Case Study 2 (Sustainability)

Part II: Small Group Discussions

• Discussion Guidance

  • Consider different areas of concerns and strategies to resolve concerns. For example: 

    • What should the process look like?

    • Who should be heard in the process? (e.g.,  different stakeholders, including academia, practitioners, policy-makers, impacted communities, etc.)

    • Where does trust come into play? What other ethical considerations might arise?

    • How would you reconcile technical and nontechnical perspectives?

Part III: Discussion and Share

Case Study 1: Robotics to Support Healthcare

Issue: Many hospitals and healthcare organizations are short-staffed and struggle to meet the needs of patients.

Example Case: The Moxi healthcare support robots from Diligent Robotics have been deployed in numerous hospitals to assist nurses and other clinical.

Seed Questions: What are the key ethical and privacy-related issues that arise from these types (avoiding collisions with patients and staff, ensuring correct and timely deliveries, gaining trust from humans, etc.)? 

Additional Resources:

• Diligent Robotics

Case Study 2: AI & Sustainability 

Reference Doc

Issue: In responsible AI efforts, how should we be balancing environmental considerations? 

Example Case: What is the tradeoff between the energy consumption and LLMs and other generative AI throughout their lifecycle (development, deployment, and usage), and how could these be addressed? 

Seed Questions: What are the considerations at different stages of the AI lifecycles for LLMs and GenAI? Is it worth exploring? Should the carbon footprint of LLMs be monitored? Would it result in changes of the model? Would clients be open to discussing? Would individuals (users) like to know? 

Additional Resources:

• AI and ESG: Understanding the Environmental Impact of AI and LLMs  

• FAccT '22: Measuring the Carbon Intensity of AI in Cloud Instances

• MIT Technology Review: AI’s carbon footprint is bigger than you think

• WEF: What would it take to make AI ‘greener’?