AI Governance & Ethics
This badge covers the specific governance artifacts and regulatory tests used to oversee AI systems: EU AI Act risk classification, NIST AI Risk Management Framework profiles, ISO/IEC 42001 management system controls, algorithmic impact assessments, model cards, datasheets for datasets, and post-market monitoring triggers. Students work with those documents directly rather than discussing governance ideas at a distance. The courses move from regulatory mapping to risk-control design and then to audit practice, including disparity-based bias checks, red-team planning, and AI incident response under the EU AI Act’s serious incident reporting duties. Each course uses concrete system examples such as credit scoring, resume screening, healthcare triage, and public-sector benefit decisions.
您将学到
课程纲要
Covers the EU AI Act’s prohibited, high-risk, limited, and minimal risk tiers, extraterritorial scope, and conformity obligations alongside the OECD AI Principles and ISO/IEC 42001 management system requirements. Students classify example AI systems and map which governance controls attach to each tier.
- EU AI Act Risk Tiers and Extraterritorial Trigger 2.5h
- EU AI Act Conformity Obligations for High-Risk Tier 2.5h
- OECD AI Principles and ISO/IEC 42001 Management System Integration 2.5h
Teaches structured impact assessment for automated decisions, including NIST AI RMF profiles, Canadian Directive on Automated Decision-Making questionnaires, and GDPR data protection impact assessment intersections. Students build control recommendations for bias, explainability, human review, and monitoring failures.
- Foundations of Algorithmic Impact Assessment: Frameworks and Questionnaires 2.5h
- Designing Risk Controls for Bias, Explainability, and Human Review 3.0h
- Monitoring Failures and Continuous Risk Control Adaptation 2.5h
Works through model card and datasheet review, disparity-metric checks, and post-deployment monitoring triggers. Students prepare an audit finding, run a structured red-team scenario, and draft an AI incident response notification.
- Model Card and Datasheet Review: Building the Audit Evidence Base 2.5h
- Disparity-Metric Checks and Bias Thresholds in Production Models 3.0h
- Structured Red-Teaming and AI Incident Response Notification 3.5h
职业相关性
A graduate can sit in a model risk or compliance review and request the artifact that should already exist — an EU AI Act risk classification memo, an AI RMF Map-Measure-Manage profile, or a model card with reported bias metrics — then check whether it is complete against the governing framework. This is directly usable by compliance officers, privacy counsel, model risk managers, AI policy leads, and vendor due-diligence teams in regulated financial, insurance, healthcare, and public-sector organizations. They are equipped to convert an audit finding into a concrete control change, such as adding a human review checkpoint or a post-deployment drift monitor, rather than stopping at a generic “AI risk” note.
建议背景知识
For compliance, legal, risk, and policy professionals who already read regulations, manage control frameworks, or draft vendor contracts. No machine learning engineering background is required, but familiarity with basic model lifecycle terms such as training data, inference, and model output helps.
课程详情
| 奖项 | Digital Badge |
| 时长 | 6 个月 |
| 难度 | Intermediate |
| 课程 | 3 |
| 语言 | English (UI in 4 languages) |
凯登学院是由 Replatform(一家于香港注册成立的公司)运营的私营全线上专业培训机构。我们并非大学、学院或经认证的教育机构。
我们未持有任何国家或地区的教育牌照、政府批准或认证,包括美国、香港、中国大陆或欧盟。
我们所颁发的证书及文凭仅由本公司作为私营企业自行颁发,并未获得任何政府、大学或官方资格认证框架所承认。这些证书不可用于:
- 转换为任何大学或学院的学分
- 移民或签证申请
- 政府就业或公务员任用用途
- 任何司法管辖区的专业牌照要求
我们的课程仅供个人及专业发展之用。雇主是否认可本学院颁发的证书,完全取决于雇主自行决定,我们不对职业发展、薪资变化或就业前景作任何保证。
本证书由香港私营企业Replatform Limited(交易名为Caden Academy)颁发,不受任何政府机构或认证机构认可,不可用于中国大陆境内的学历认证或政府就业用途。