Ethical foundations of AI

Fairness in AI ensures that AI systems treat everyone equally, without bias against any gender, race, or ethnicity. It aims to prevent discrimination and promote equality.

Example - Application Tracking Systems (ATS): If an ATS used in hiring favors certain demographics because of its training data, it's crucial to address this bias. Ensuring fairness in such AI models helps create a more inclusive society.

Explainability is about making AI's decision-making transparent, so users can understand and trust how it reaches its conclusions.

Example - AI in Finance: For an AI deciding on loan approvals, explainability means it can show which factors influenced its decision, helping applicants understand or contest the outcome.

Transparency means openly sharing how AI systems work, from data use to decision-making processes, fostering trust and informed user interaction.

Example - AI in Social Media: When social media platforms explain how their algorithms tailor user feeds, they promote transparency, empowering users to better control their online experience.

Robustness is about making AI reliable in unforeseen conditions, ensuring it performs well even when faced with challenges like data changes or attacks.

Example - Facial Recognition: A robust facial recognition system accurately identifies users under various conditions, like changes in lighting or expression, minimising errors and boosting dependability.

Privacy focuses on protecting personal data from misuse or unauthorized access, emphasizing the responsible management of sensitive information.

Example - Fitbit watches: Privacy in devices like Fitbit ensures users' health data is securely stored and accessed responsibly, fostering trust in how their information is handled.