Transparency
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.
Think Like a Software Engineering Manager
Transparency is the open sharing of relevant information, reasoning, and decisions with affected people.
- How it is used
- Supports trust, alignment, and informed work when engineering teams face changing priorities, performance conversations, cross-functional decisions, or departures.
- How deep it goes
- Develops practical communication habits and disclosure boundaries through situations and suggested actions, without prescribing a single formal transparency process.
- Examples
- Uses a reassigned engineer, a departure that unsettles colleagues, and documented changes to a cross-team project plan.
- What you can do after
- Practice/perform it by explaining relevant decisions, inviting feedback, documenting shared plans, and communicating changes without revealing protected details.
Read how Think Like a Software Engineering Manager explains transparency →
The Art of AI Product Development
Transparency is the clear communication of an AI system's capabilities, limitations, behavior, and relevant decision factors.
- How it is used
- Helps users judge AI outputs and limitations while supporting trust and adoption in consequential workflows.
- How deep it goes
- Develops practical interface and communication choices around explanations, uncertainty, and feedback, without specifying a technical explanation method.
- Examples
- Uses an unexplained marketing-budget recommendation and a demand forecast with visible prediction factors and uncertainty.
- What you can do after
- Diagnose/debug with it by spotting unexplained outputs, unclear limits, and missing opportunities for users to challenge predictions.
Read how The Art of AI Product Development explains transparency →
Generative AI in Action
Transparency is the degree to which an AI system's operation, information sources, limitations, and decision processes are made visible to relevant people.
- How it is used
- Positions transparency as a safeguard in GenAI deployment, helping users understand AI involvement, model limitations, and referenced information.
- How deep it goes
- Details disclosures, model documentation, and verifiable evaluations, but does not specify a technical method for explaining individual model decisions.
- Examples
- Illustrates transparency through model cards, source citations, AI-role disclosure, and recorded evaluation criteria, methods, and results.
- What you can do after
- Practice/perform user disclosure, source attribution, model-document review, and transparent evaluation during deployment.
Introduction to Generative AI, Second Edition
Transparency is the disclosure of data, methods, evaluations, and system behavior needed to understand and assess an artificial-intelligence system.
- How it is used
- Frames how people assess generative AI performance, personal-data use, and accountability.
- How deep it goes
- Distinguishes dataset, model, and system-level disclosure, while leaving implementation details of interpretability and debugging unspecified.
- Examples
- Uses a code-generation leaderboard, inspectable assistant memory, companion-chatbot changes, and student-data governance to show different disclosure needs.
- What you can do after
- Evaluate/understand AI disclosures by asking what data, evaluations, behavior, and controls are visible—and what remains uncertain.
Read how Introduction to Generative AI, Second Edition explains transparency →
Delivering Machine Learning Projects: From design to deployment
Transparency is the degree to which a model's reasoning and outputs can be understood and inspected by people.
- How it is used
- Frames transparency as a user-story design force that constrains classifier selection and the interface through which people inspect decisions.
- How deep it goes
- Develops model-selection trade-offs and interface options, but stops short of implementing or validating an explanation method.
- Examples
- Uses AlphaFold 2 structure renderings, activation maps, and an energy-use decision log as ways to inspect otherwise difficult-to-explain decisions.
- What you can do after
- Evaluate/understand it by weighing explainability against model performance and planning what users can inspect when an opaque model is selected.
Read how Delivering Machine Learning Projects: From design to deployment explains transparency →
First-Party Data Activation
Transparency is the clear disclosure of what data is collected, how it is used, and why it is used.
- How it is used
- Builds trust in first-party data collection by making data practices understandable to customers.
- How deep it goes
- Defines disclosure and gives customer-facing practices, but stops short of technical implementation instructions.
- Examples
- Uses a purchase-history recommendation statement, data access logs, customer dashboards, privacy toggles, and preference centers.
- What you can do after
- Practice/perform it through clear disclosures and accessible controls that help customers understand data use and revise their choices.
Read how First-Party Data Activation explains transparency →
Collaborative Software Design: How to facilitate domain modeling decisions
Transparency is the degree to which an intervention's purpose, method, and influence are openly disclosed to the people affected by it.
- How it is used
- Guides ethical use of nudges by relating awareness and choice to decision-making and collaborative modeling.
- How deep it goes
- Distinguishes transparent and nontransparent nudges through quadrant guidance and a facilitation example, without developing a general disclosure checklist.
- Examples
- Uses a BigScreen facilitation exchange and a quadrant figure contrasting transparent Type II with nontransparent Type I nudges.
- What you can do after
- Evaluate nudges by awareness and choice, then invite reflection with questions about evidence that could change a participant’s mind.
Lead Developer Career Guide
The quality of making project information, decisions, progress, and status visible and understandable to relevant people.
- How it is used
- Builds trust and manages expectations among clients, mentees, and development teams through visible information and open communication.
- How deep it goes
- Provides practical communication tactics and workplace examples but stops short of formal systems or detailed implementation guidance.
- Examples
- Uses client dashboards, weekly meetings, progress reports, code-review comments, and contextual deployment updates.
- What you can do after
- Apply it by giving timely updates, explaining constraints, surfacing risks, and creating feedback loops for shared understanding.
Read how Lead Developer Career Guide explains transparency →
Starting Data Analytics with Generative AI and Python
Transparency is the degree to which a system’s decisions, processes, and relevant information are made understandable and inspectable.
- How it is used
- Uses transparency to govern responsible AI deployment, communicate uncertainty, and support accountability across model use and results.
- How deep it goes
- Explains documentation, monitoring, verification, and disclosure practices, but stops short of technical interpretability methods.
- Examples
- Uses AI policies, model documentation, performance reviews, stakeholder reports, verification records, and disclosed human overrides.
- What you can do after
- Evaluate AI outputs by tracing their sources, verification, uncertainty, limitations, and human interventions before communicating decisions.
Read how Starting Data Analytics with Generative AI and Python explains transparency →
Writing for Developers: Blogs that get read
The quality of communicating intentions, scope, limitations, and expected content openly and clearly.
- How it is used
- Uses openness to frame product-oriented engineering posts around reader value, technical perspective, and clear product disclosure.
- How deep it goes
- Provides an editorial principle, a detailed anti-selling checklist, and introduction guidance rather than technical implementation.
- Examples
- Uses a TigerBeetle distributed-consistency example, reader-payoff disclosures, source-code expectations, and an anti-selling checklist.
- What you can do after
- Evaluate and revise engineering posts so readers understand their payoff, product context, evidence, and limits before committing to them.
Read how Writing for Developers: Blogs that get read explains transparency →
Web Components in Action
Transparency is the degree to which a rendered color allows the background behind it to show through.
- How it is used
- Uses transparency as a continuously adjustable dimension in an always-visible Web Components color picker.
- How deep it goes
- Explains visual gradients, hexadecimal alpha, attribute updates, and dragging, while deferring Edge compatibility.
- Examples
- Demonstrates a checkered transparency background, a fading blue slider, hexadecimal alpha, and a draggable custom element.
- What you can do after
- Operate/configure a reusable transparency slider through color and value attributes, then adjust it by dragging.
Inside AI
Transparency is the property of making a system's data, processes, or decision basis sufficiently visible for scrutiny.
- How it is used
- Guides the validation and deployment of AI systems whose decisions affect individuals or society.
- How deep it goes
- Explains visible inner workings, decision rationale, and plain-language explanations, but does not detail implementation methods.
- Examples
- Uses incarceration, probation, lending, criminal justice, and FDA-style certification as concrete high-stakes examples.
- What you can do after
- Evaluate AI systems for understandable reasoning, bias, reliability, fairness, privacy, and suitability for consequential decisions.
The Complete Obsolete Guide to Generative AI
Transparency is the availability of information about a system's design, operation, decision-making, and effects sufficient for external understanding or scrutiny.
- How it is used
- Positions transparency as a safeguard for trustworthy, accountable AI decisions and ethical startup adoption.
- How deep it goes
- Defines explainability and interpretability through decision factors, reasoning, bias detection, and high-stakes applications, while treating advances as ongoing research.
- Examples
- Uses credit scoring, legal contract analysis, educational assessment, startup AI operations, and AGI governance as contexts.
- What you can do after
- Evaluate/understand AI outputs by seeking explanations, checking influential factors, and identifying possible bias or errors.
Read how The Complete Obsolete Guide to Generative AI explains transparency →
Building Reliable AI Systems
The degree to which an AI system's behavior, decisions, and relevant processes can be understood or inspected.
- How it is used
- Makes RAG answers and agent actions inspectable so users can verify sources, follow decisions, and flag needed corrections.
- How deep it goes
- Explains source references and visible agent steps, then outlines a policy-answer workflow; stops short of a dedicated transparency evaluation procedure.
- Examples
- Uses a health-care policy assistant returning document sources and an agent interface displaying steps beside citations.
- What you can do after
- Apply it via source-bearing RAG answers and visible agent steps to help users inspect responses and identify retrieval gaps.
Read how Building Reliable AI Systems explains transparency →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
The degree to which a model's operation, predictions, and feature reliance can be understood and assessed by people.
- How it is used
- Guides model selection by balancing stakeholder understanding and feature assessment alongside simplicity and efficacy.
- How deep it goes
- Separates transparency into explainability and feature importance, then compares gradient boosting with deep learning.
- Examples
- Uses Airbnb price prediction and house time-on-market prediction to compare practical model transparency.
- What you can do after
- Evaluate model transparency by explaining behavior and identifying influential features for stakeholders or regulators.
Read how Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI explains transparency →
How Large Language Models Work
Transparency is the degree to which an AI system's identity, development, modification, capabilities, and limitations are made discoverable to its users.
- How it is used
- Uses discoverable system information and automated-response disclosure to set expectations and support trust.
- How deep it goes
- Provides practical disclosure questions but stops short of prescribing a universal method for building trust.
- Examples
- Uses model disclosures, simulated experts, consent questions, verification paths, and automated-response notices.
- What you can do after
- Apply transparency by disclosing system identity, modifications, limitations, automation, and verification paths.
Read how How Large Language Models Work explains transparency →
Think Like a CTO
An organizational communication practice of making relevant information understandable and available to affected people while respecting confidentiality constraints.
- How it is used
- Uses communication to connect teams with business context, coordinate reporting, and build confidence in leadership decisions.
- How deep it goes
- Explains timing, confidentiality, contextualization, reporting cadence, and promotion announcements without prescribing a universal disclosure rule.
- Examples
- Uses confidential executive information, technology roadmaps, reporting routines, and promotion announcements as practical situations.
- What you can do after
- Apply it via communication routines that select appropriate timing, provide context, set reporting expectations, and explain consequential decisions.
Math for Web Design
Transparency is the visual property that allows underlying colors or content to show through a foreground element.
- How it is used
- Explains how layered colors produce visible transparency in browser-rendered graphics.
- How deep it goes
- Derives per-channel compositing with normalized RGB values and a worked calculation, while stopping short of other blend modes.
- Examples
- Uses semitransparent blue over red and CSS examples for alpha-enabled rgb() and hexadecimal colors.
- What you can do after
- Apply transparency via CSS rgb() or hexadecimal alpha notation and predict the resulting mixed color.
Financial AI in Practice
Transparency is the quality of making a model's processes, inputs, or outputs sufficiently understandable to support scrutiny and accountability.
- How it is used
- Supports scrutiny, compliance, and trust in financial AI by making outputs and decision processes reviewable.
- How deep it goes
- Demonstrates source displays and decision logs, but does not explain model internals.
- Examples
- Uses a See Sources feature, output fairness audits, and an investment decision log showing reasoning, weights, and flags.
- What you can do after
- Diagnose/debug with it by tracing logged inputs, reasoning, recommendations, flags, and outcomes.
Related concepts
- interpretabilitycontrasts with · Interpretability is the degree to which people can understand a model’s behavior, outputs, predictions, or the factors that produce them.
- alpha channelconfigured by · An alpha channel is a color or image component that specifies the opacity or transparency of each color or pixel.
- retrieval-augmented generationproduced by · Retrieval-augmented generation is a technique that retrieves relevant information from external sources and supplies it as context to a generative model when producing an output.
- Digital Services Actconfigured by · The Digital Services Act is a European Union regulation that establishes obligations for online intermediary services and platforms, including requirements concerning transparency, accountability, and risk management.
- encapsulationcontrasts with · Encapsulation is a software design principle that contains state and implementation details behind a controlled interface, limiting direct access from outside.
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.