AI decision-making

AI decision-making refers to processes where artificial intelligence (AI) systems provide guidance or input to help humans make choices. In some of these processes, AI plays a more supportive role where outputs are reviewed and analyzed by humans prior to arriving at any conclusion. But as AI systems evolve, their role in decision-making processes can become more significant, introducing novel ethical concerns.

AI decision-making is in use across many fields and applications including:

• Healthcare: AI assists medical professionals in diagnosing diseases by analyzing lab results and patient histories.
• Finance: AI algorithms assess creditworthiness and are used in fraud detection.
• Politics: AI agents influence formal and informal deliberative processes like citizen assemblies and the spreading of ideas in digital spaces.
• Sales: AI tools within customer data platforms can automate prospecting based on buyer intent.
• Policing: Data fusion systems, leveraging AI, enable police to develop sophisticated hypotheses about who may or may not have committed a crime.

Explore the resources below for more on the ethical challenges connected to AI decision-making.

1. Should humans always have the ability to override a decision made by AI regardless of context?
2. Under what kinds of conditions can more autonomy be given to AI to make decisions without human oversight, if any?
3. Should AI systems be required to explain their decisions? Why or why not?
4. How should responsibility be distributed between the developers, the users, or the AI itself when an AI system is involved in a decision that culminates in a bad outcome?
5. What ethical guidelines should govern the use of AI systems for decision-making? How do those guidelines change based on context?
6. How can we improve the fairness of AI decision-making?
7. Should development of AI decision-making systems be regulated?