book

Responsible AI in the Enterprise

Name: Responsible AI in the Enterprise
ISBN: 9781803230528

by Adnan Masood, Heather Dawe

July 2023

Intermediate to advanced

318 pages

9h 16m

English

Packt Publishing

Read now

Unlock full access

Responsible AI in the Enterprise
ForewordForeword 2ContributorsAbout the authorsAbout the reviewers
Preface
Who this book is forEssential chapters tailored to distinct AI-related positionsWhat this book coversTo get the most out of this bookDownload the example code filesConventions usedGet in touchShare Your ThoughtsDownload a free PDF copy of this book
Part 1: Bigot in the Machine – A Primer
Chapter 1: Explainable and Ethical AI Primer
The imperative of AI governanceKey terminologiesExplainabilityInterpretabilityExplicabilitySafe and trustworthyFairnessEthicsTransparencyModel governanceEnterprise risk management and governanceTools for enterprise risk governanceAI risk governance in the enterprisePerpetuating bias – the network effectTransparency versus black-box apologetics – advocating for AI explainabilityThe AI alignment problemSummaryReferences and further reading
Chapter 2: Algorithms Gone Wild
AI in hiring and recruitmentFacial recognitionBias in large language models (LLMS)Hidden cost of AI safety – low wages and psychological impactAI-powered inequity and discriminationPolicing and surveillanceSocial media and attention engineeringThe environmental impactAutonomous weapon systems and militaryThe AIIDSummaryReferences and further reading
Part 2: Enterprise Risk Observability Model Governance
Chapter 3: Opening the Algorithmic Black Box
Getting started with interpretable methodsThe business case for explainable AITaxonomy of ML explainability methodsShapley Additive exPlanationsHow is SHAP different from Shapley values?A working example of SHAPLocal Interpretable Model-Agnostic ExplanationsA working example of LIMEFeature importanceAnchorsPDPsCounterfactual explanationsSummaryReferences and further reading
Chapter 4: Robust ML – Monitoring and Management
An overview of ML attacks and countermeasuresModel and data securityPrivacy and complianceAttack prevention and monitoringEthics and responsible AIThe ML life cycleAdopting an ML life cycleMLOps and ModelOpsModel driftData driftConcept driftMonitoring and mitigating drift in ML modelsSimple data drift detection using Python data drift detectorHousing price data drift detection using EvidentlyAnalyzing data drift using Azure MLSummaryReferences and further reading
Chapter 5: Model Governance, Audit, and Compliance
Policies and regulationsUnited StatesEuropean UnionUnited KingdomSingaporeUnited Arab EmiratesToronto Declaration – protecting the right to equality in MLProfessional bodies and industry standardsMicrosoft’s Responsible AI frameworkIEEE Global Initiative for Ethical Considerations in AI and Autonomous SystemsISO/IEC’s standards for AIOECD AI PrinciplesThe University of Oxford’s recommendations for AI governancePwC’s Responsible AI Principles/ToolkitAlan Turing Institute guide to AI ethicsTechnology toolkitsMicrosoft FairlearnIBM’s AI Explainability 360 open source toolkitCredo AI Lens toolkitPiML – the integrated Python toolbox for interpretable MLFAT Forensics – algorithmic fairness, accountability, and transparency toolboxAequitas – the Bias and Fairness Audit ToolkitAI trust, risk, and security managementAuditing checklists and measuresDatasheets for datasetsModel cards for model reportingSummaryReferences and further reading
Chapter 6: Enterprise Starter Kit for Fairness, Accountability, and Transparency
Getting started with enterprise AI governanceAI STEPS FORWARD – AI governance frameworkImplementing AI STEPS FORWARD in an enterpriseThe strategic principles of AI STEPS FORWARDAI STEPS FORWARD in enterprise governanceThe AI STEPS FORWARD maturity modelRisk management in AI STEPS FORWARDMeasures and metrics of AI STEPS FORWARDAI STEPS FORWARD – taxonomy of componentsSalient capabilities for AI GovernanceThe indispensable role of the C-suite in fostering responsible AI adoptionAn AI Center of ExcellenceThe role of internal AI boards in enterprise AI governanceHealthcare systemsRetail and e-commerce systemsFinancial servicesPredictive analytics and forecastingCross-industry applications of AIEstablishing repeatable processes, controls, and assessments for AI systemsEthical AI upskilling and educationSummaryReferences and further reading

Part 3: Explainable AI in Action
Chapter 7: Interpretability Toolkits and Fairness Measures – AWS, GCP, Azure, and AIF 360
Getting started with hyperscaler interpretability toolkitsGoogle Vertex Explainable AIModel interpretability in Vertex AI – feature attribution and example-based explanationsIntegration with Google Colab and other notebooksSimplified deploymentExplanations are comprehensive and multimodalAWS Sagemaker ClarifyAzure Machine Learning model interpretabilityAzure’s responsible AI offeringsResponsible AI scorecardsOpen source offerings – the responsible AI toolboxOpen source toolkits and lensesIBM AI Fairness 360Aequitas – Bias and Fairness Audit ToolkitPETsDifferential privacyHomomorphic encryptionSecure multiparty computationFederated learningData anonymizationData perturbationSummaryReferences and further reading
Chapter 8: Fairness in AI Systems with Microsoft Fairlearn
Getting started with fairnessFairness metricsFairness-related harmsGetting started with FairlearnSummaryReferences and further reading
Chapter 9: Fairness Assessment and Bias Mitigation with Fairlearn and the Responsible AI Toolbox
Fairness metricsDemographic parityEqualized oddsSimpson’s paradox and the risks of multiple testingBias and disparity mitigation with FairlearnFairness in real-world scenariosMitigating correlation-related biasThe Responsible AI ToolboxThe Responsible AI dashboardSummaryReferences and further reading
Chapter 10: Foundational Models and Azure OpenAI
Foundation modelsBias in foundation modelsThe AI alignment challenge – investigating GPT-4’s power-seeking behavior with ARCEnterprise use of foundation models and bias remediationBiases in GPT3Azure OpenAIAccess to Azure OpenAIThe Code of ConductAzure OpenAI Service content filteringUse cases and governanceWhat not to do – limitations and potential risksData, privacy, and security for Azure OpenAI ServiceAI governance for the enterprise use of Azure OpenAIGetting started with Azure OpenAIConsuming the Azure OpenAI GPT3 model using the APIAzure OpenAI Service modelsCode generation modelsEmbedding modelsSummaryReferences and further reading
Index
Why subscribe?
Other Books You May EnjoyPackt is searching for authors like youShare Your ThoughtsDownload a free PDF copy of this book

Content preview from Responsible AI in the Enterprise

7 Interpretability Toolkits and Fairness Measures – AWS, GCP, Azure, and AIF 360

Throughout this book, we have discussed the rationale for why responsible AI and AI governance are increasingly becoming critical disciplines for enterprises that wish to leverage AI. We also provided an overview of the methods that can be used to test that the machine learning models underpinning AI are safe, fair, and fit for purpose, along with introducing an AI assurance framework – AI STEPS FORWARD.

Leading cloud AI providers – the hyperscalers (namely AWS, Google, and Microsoft) – have recognized the need for AI explainability, and each has developed explainability toolkits designed to be used with its respective ML/AI development and MLOps environments. At ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Managing AI in the Enterprise: Succeeding with AI Projects and MLOps to Build Sustainable AI Organizations

Publisher Resources

ISBN: 9781803230528

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Responsible AI in the Enterprise

by Adnan Masood, Heather Dawe

7

Interpretability Toolkits and Fairness Measures – AWS, GCP, Azure, and AIF 360

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.