Course Description
In these lessons on Professional Responsibility and Human Control of AI, you will learn about the broad principle of human control of technology and how it intersects with AI system development. You will also see how the designers’ responsibility is linked to AI system governance and multistakeholder collaboration to enable valuable feedback about the system design and its impact on various stakeholder groups.
Human control of technology is a more general set of principles that govern technology systems in general. By building automated decision systems, we shift the control of those decisions from a human to an AI system. As such, it’s essential to consider why and when humans should review decisions of an AI system. In these lessons, you’ll explore the principle of capacity for human intervention, including how humans can make the decision of what actions to delegate to a technology system. You’ll also understand why AI systems should allow humans to intervene in the decisions and actions made by the system.
Because professionals who are working in the design, creation, and implementation of AI systems have a unique vantage point, they can see an AI system in a way that others cannot. You’ll learn how the responsibility of designers as a principle is linked to the governance of AI systems. You’ll also get an overview of the responsibility of designers of an AI system, including long-term impacts of an AI system.
Finally, you’ll learn about multistakeholder collaboration as a way to hear concerns about the system design and its impact on various stakeholder groups, as well as the importance of international collaboration on the development of AI systems.
What You'll Learn
- Understand the broad principle of human control of technology and how it intersects with AI system development
- Explain why and when humans should review the decisions of an AI system
- Apply the principle of capacity for human intervention, including deciding which actions to delegate to a technology system
- Describe the unique responsibilities of designers and how that responsibility links to AI system governance
- Assess the long-term impacts of an AI system from the designer's vantage point
- Describe multistakeholder collaboration as a way to gather feedback on system design and its impact on stakeholder groups
Key Takeaways
- Human control of technology is a general set of principles that govern technology systems, and it intersects with how AI systems are developed.
- Building automated decision systems shifts control of those decisions from a human to an AI system, making it essential to consider why and when humans should review the system's decisions.
- AI systems should allow humans to intervene in the decisions and actions the system makes, and humans decide which actions to delegate to a technology system.
- Professionals working in the design, creation, and implementation of AI systems have a unique vantage point, so the responsibility of designers is linked to the governance of AI systems and the system's long-term impacts.
- Multistakeholder and international collaboration helps surface concerns about the system design and its impact on various stakeholder groups.
Frequently Asked Questions
What does this course cover?
It covers professional responsibility and human control of AI, including human review of decisions, capacity for human intervention, the responsibility of designers, long-term impacts of AI systems, and multistakeholder collaboration.
Who is this course for?
It is aimed at professionals working in the design, creation, and implementation of AI systems, who have a unique vantage point on how an AI system works.
What skills will I gain?
The course develops skills in accountability, professional responsibility, artificial intelligence and AI systems, AI development, and defining roles and responsibilities.
What are the main learning objectives?
To understand AI's relationship to human control of technology, understand the unique responsibilities of designers, and describe multistakeholder collaboration for ethical AI systems.
Why is human intervention emphasized in this course?
Because building automated decision systems shifts control from humans to AI systems, the course explains why AI systems should allow humans to intervene and how humans decide which actions to delegate to a technology system.
