How deeply does How Large Language Models Work cover transparency?

Provides practical disclosure questions but stops short of prescribing a universal method for building trust.

Transparency

As explained in 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.

Transparency makes an AI system’s identity, development, modifications, capabilities, and limits discoverable. It also clearly tells people when automated responses—not a human—are in control. section 8.4.1

Useful transparency answers the questions a skeptical user or auditor would reasonably ask:

Which model was used, and how was it modified? section 8.4.1
Does the system imitate a person or credentialed professional, and did that person or an equivalent expert consent or approve its effectiveness? section 8.4.1
What can the system do, what can it not do, and how can users verify the answers? section 8.4.1

A transparent interaction

A health assistant identifies itself as an automated AI, names its underlying model at a high level, states that it has limited capability, and provides a way to inspect supporting information. This sets more realistic expectations than presenting the system as a human expert. section 8.4.1 section 8.4

How How Large Language Models Work covers it

liveBook manning.com

Read How Large Language Models Work on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Think Like a Software Engineering ManagerTransparency is the open sharing of relevant information, reasoning, and decisions with affected people.
liveBook manning.com

The Art of AI Product DevelopmentTransparency is the clear communication of an AI system's capabilities, limitations, behavior, and relevant decision factors.
liveBook manning.com

Generative AI in ActionTransparency is the degree to which an AI system's operation, information sources, limitations, and decision processes are made visible to relevant people.
liveBook manning.com

Introduction to Generative AI, Second EditionTransparency is the disclosure of data, methods, evaluations, and system behavior needed to understand and assess an artificial-intelligence system.
liveBook manning.com

Delivering Machine Learning Projects: From design to deploymentTransparency is the degree to which a model's reasoning and outputs can be understood and inspected by people.
liveBook manning.com

First-Party Data ActivationTransparency is the clear disclosure of what data is collected, how it is used, and why it is used.
liveBook manning.com

Collaborative Software Design: How to facilitate domain modeling decisionsTransparency is the degree to which an intervention's purpose, method, and influence are openly disclosed to the people affected by it.
liveBook manning.com

Lead Developer Career GuideThe quality of making project information, decisions, progress, and status visible and understandable to relevant people.
liveBook manning.com

Starting Data Analytics with Generative AI and PythonTransparency is the degree to which a system’s decisions, processes, and relevant information are made understandable and inspectable.
liveBook manning.com

Writing for Developers: Blogs that get readThe quality of communicating intentions, scope, limitations, and expected content openly and clearly.
liveBook manning.com

Web Components in ActionTransparency is the degree to which a rendered color allows the background behind it to show through.
liveBook manning.com

Inside AITransparency is the property of making a system's data, processes, or decision basis sufficiently visible for scrutiny.
liveBook manning.com

The Complete Obsolete Guide to Generative AITransparency is the availability of information about a system's design, operation, decision-making, and effects sufficient for external understanding or scrutiny.
liveBook manning.com

Building Reliable AI SystemsThe degree to which an AI system's behavior, decisions, and relevant processes can be understood or inspected.
liveBook manning.com

Machine Learning for Tabular Data: XGBoost, Deep Learning, and AIThe degree to which a model's operation, predictions, and feature reliance can be understood and assessed by people.
liveBook manning.com

Think Like a CTOAn organizational communication practice of making relevant information understandable and available to affected people while respecting confidentiality constraints.
liveBook manning.com

Math for Web DesignTransparency is the visual property that allows underlying colors or content to show through a foreground element.
liveBook manning.com

Financial AI in PracticeTransparency is the quality of making a model's processes, inputs, or outputs sufficiently understandable to support scrutiny and accountability.

Compare how 19 books explain transparency →

Related concepts

interpretabilityhas part · 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.
principlekind of · A principle is a general rule, guideline, or value that guides decisions, behavior, or judgments.
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.
importance scorehas part · An importance score is a numerical value indicating an item's relative significance, relevance, or contribution within a specified context.

This explainer was generated by Manning's knowledge graph from the text of How Large Language Models Work. The section links point to the parts of the book it draws on.