Ethics
Ethics is the study and application of principles for evaluating morally acceptable conduct, decisions, and outcomes.
Delivering Machine Learning Projects: From design to deployment
Ethical considerations are judgments about whether a system's purposes, effects, and operation respect human welfare, rights, and societal values.
- How it is used
- Tests whether a proposed ML project and its data use remain appropriate for the people affected before and during development.
- How deep it goes
- Turns review into proposal checks, impact assessments, and data-history documentation; stops short of a universal ethical standard or specialist-domain questionnaire.
- Examples
- Uses unknown personal-data origins, social networking’s societal effects, and criminal prediction from appearance to show distinct risks.
- What you can do after
- Practice/perform project-level ethical review by seeking affected perspectives, assessing impacts, documenting data origins, and reconsidering development when concerns arise.
Read how Delivering Machine Learning Projects: From design to deployment explains ethics →
LLMs in Production
Principles concerning morally acceptable behavior and the consequences of technological decisions.
- How it is used
- Frames LLM bias, automated detection, and regulation as decisions with practical legal and societal consequences.
- How deep it goes
- Distinguishes moral from ethical responsibility through a thought experiment, but stops short of formal philosophical analysis.
- Examples
- Uses false-positive detection systems, arbitrary teacher enforcement, and the trolley problem to illustrate ethical consequences.
- What you can do after
- Evaluate/understand technological decisions by separating moral blame from ethical and societal consequences.
Effective Data Analysis
Ethics in science concerns principles governing responsible scientific practice and the treatment of people, animals, and society.
- How it is used
- Guides data analysis, experimentation, and scientific practice toward responsible treatment of people and reduction of harm.
- How deep it goes
- Connects ethical principles to protected data, anonymization, consent, bias, access, and misuse without providing detailed procedures.
- Examples
- Uses animal research debates, unequal access to advances, misuse of knowledge, and Amazon’s biased resume-ranking model.
- What you can do after
- Evaluate data and scientific work for stakeholder effects, potential bias, unintended consequences, and opportunities to reduce harm.
Related concepts
- GDPRconfigured by · The General Data Protection Regulation (GDPR) is a European Union regulation governing the processing of personal data and protecting individuals’ rights concerning that data.
- decision makinguses · Decision making is the process of selecting an action, outcome, or judgment from available alternatives based on information, objectives, or constraints.
- transparencyhas part · Transparency is the degree to which underlying content is visible through an element or relevant information, processes, and decisions are accessible and understandable to observers.
- intellectual propertyhas part · Intellectual property comprises creations of the mind, such as inventions, expressive works, designs, and distinctive marks, together with the legal rights that protect them.
- IEEE use case matrixproduced by · The IEEE use case matrix is a framework for relating artificial-intelligence application contexts to potentially relevant ethical concerns.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.