100+ Free CertNexus CET Practice Questions

Fundamental Concepts for Data-Driven Technology Ethics

Terminology (AI, data science, legal, privacy, ethics, bias, evaluation metrics like precision/recall/F1), ethical theories (consequentialism/utilitarianism, deontology/Kant, virtue ethics/Aristotle, contractualism, care ethics), bias types (selection, sampling, measurement, confirmation, automation, algorithmic, representation, aggregation, historical), and triggers for ethical risk review (ideation, what-if, new product, full DS/AI lifecycle, Ethics-by-Design, post-incident)

Ethical Frameworks

Common principles (privacy, accountability, safety/security, transparency/explainability, fairness/non-discrimination, human control, professional responsibility, human values), framework selection by industry/regulator/risk, regulations and standards (NIST AI RMF 1.0 GOVERN/MAP/MEASURE/MANAGE, EU AI Act risk tiers, GDPR Article 22, ISO/IEC 24368, IEEE 7000-2021, IEEE 7001/7002/7003, OECD AI Principles, UNESCO 2021 Recommendation, Asilomar AI Principles, US EO 14110, NYC LL 144, Illinois AI Video Interview Act, Colorado SB24-205, Privacy by Design 7 principles), and ethics-vs-business conflicts (data minimization vs need, performance vs explainability, compliance vs cost, transparency vs IP, ethics washing, efficiency vs collateral, AI proliferation vs democratization, IoT/big data vs concentration of power, fair competition vs hegemony, moral relativism vs evidence-based)

Risk Identification and Mitigation

Privacy risks and mitigations (differential privacy, k-anonymity, l-diversity, federated learning, secure aggregation, dark patterns), accountability risks (responsibility gap, AI Impact Assessments, DPIA under GDPR Article 35, named ownership, audit trails), transparency/explainability risks (model cards, datasheets for datasets, AI fact sheets, system cards, SHAP, LIME, counterfactual explanations, surrogate models, Captum), fairness/non-discrimination risks (demographic parity, equal opportunity, equalized odds, predictive parity, calibration, four-fifths rule, impossibility theorem, disaggregated monitoring), and safety/security risks (Security-by-Design, NIST CSF 2.0, OWASP AI Security and Privacy Guide, MITRE ATLAS, OWASP Top 10 for LLM Applications, prompt injection, data poisoning, model evasion, red-teaming)

Communication

Internal stakeholder communication of ethical risks (tying ethics to business impact, executive briefings, ethics one-pagers, RACI for ethics owners, incentives that reward risk identification), and external communication (marketing/PR avoiding ethics washing, brand defense after incidents, coordinated media inquiry response, corporate ESG/responsible-AI reporting, public ethical philosophy, user disclosures aligned with EU AI Act transparency obligations)

Organizational Policy and Governance

Fostering an ethical culture (visible leadership, layered training and workshops, aligned incentives, ethics board with charter and authority, escalation channels, resourcing), policy considerations (fair competition, open data, privacy, IP, fairness/non-discrimination, legal/regulatory, human rights, accountability, transparency, animal rights, safety/reliability, environmental, economic, workforce), code of ethics development (ACM Code of Ethics 2018, IEEE Code of Ethics, PMI Code, ICAEW), ethical policy lifecycle (draft, ratify, communicate, train, review, update), and evaluating program effectiveness (sentiment analysis, employee ethics surveys, periodic health checks, violation severity matrix, industry best-practice benchmarking)