Module 2: AI Impacts & Responsible AI, BoK II.A
Seven ethical issues and three foundational controls
The Seven ethical issues responsible AI must address - lawfulness, safety, bias protection, transparency, choice, human intervention, security - and the Three foundational controls that mitigate ethical risk: ethical principles, an Oversight body, and policies & procedures.
First the issues responsible AI must address, then the minimum control set that mitigates ethical risk.
The seven key ethical issues:
- Lawfulness → operate within existing laws and regulations protecting rights and norms.
- Safety → prioritise the safety and well-being of people and the environment.
- Bias protection → minimise and eliminate biases → equitable outcomes across populations.
- Transparency → clear explanations of how decisions are made and what data is used.
- Choice → people stay informed and empowered over how their personal data is collected, used and shared to develop AI.
- Human intervention → option to request human oversight in AI decisions affecting legal rights or well-being.
- Security → organisations are accountable for securing the AI they build and use against vulnerabilities and threats.
The three foundational controls:
- Ethical principles → adopt and adhere to AI ethical principles → mitigates bias, privacy violations, opacity from the start.
- Oversight body → a cross-functional, demographically diverse body reviewing higher-risk use cases in ethical grey areas → regular training keeps guidance informed.
- Policies & procedures → assess and enhance policies for unfair bias, disparate impact, privacy, cybersecurity, data governance → establish metrics to evaluate effectiveness and adjust proactively.
Key terms - quick answers
What is “Seven ethical issues”?
Lawfulness, safety, bias protection, transparency, choice, human intervention, security.
What is “Three foundational controls”?
Ethical principles, an oversight body, and policies & procedures (with effectiveness metrics).
What is “Oversight body”?
A cross-functional, demographically diverse body reviewing higher-risk AI use cases in ethical grey areas.
Sources and study method
This independent lesson uses active recall, spaced retrieval and scenario practice. Read the full study method.