+254 721 331 808    training@upskilldevelopment.com

Home / Monitoring and Evaluation

Generative AI Governance and Risk Management for Evaluation Institutions Training Course

Build the expertise to govern generative AI confidently across evaluation institutions. Learn to manage ethical, legal, data, security, and evidence-quality risks through practical policies, risk registers, human oversight, vendor controls, incident response, and measurable assurance systems that strengthen trust and accountability.
Duration: 10 Days Classroom & Online In-House Available Certificate on Completion

Training Registration & Schedule

Compare training options, fees and available dates, then register for your preferred session.

Online Training Registration

Training ModePlatformDurationFee per ParticipantRegistration
LIVE ONLINEInstructor-led training Zoom / Google MeetAttend from anywhere 10 days USD 1,740 Register Online →

Classroom / On-site Training Schedule

15 upcoming sessions

Swipe horizontally to see all columns →

Training DatesLocationTraining ModeFee per ParticipantRegistration
07 Dec 2026 – 18 Dec 202610 training days Mombasa CLASSROOM USD 3,400 Register Now →
21 Dec 2026 – 01 Jan 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
18 Jan 2027 – 29 Jan 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
18 Jan 2027 – 29 Jan 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
15 Feb 2027 – 26 Feb 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
15 Feb 2027 – 26 Feb 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
15 Mar 2027 – 26 Mar 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
15 Mar 2027 – 26 Mar 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
19 Apr 2027 – 30 Apr 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
19 Apr 2027 – 30 Apr 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
17 May 2027 – 28 May 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
17 May 2027 – 28 May 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
21 Jun 2027 – 02 Jul 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
21 Jun 2027 – 02 Jul 202710 training days Mombasa CLASSROOM USD 3,400 Register Now →
19 Jul 2027 – 30 Jul 202710 training days Nairobi CLASSROOM USD 2,900 Register Now →
Need registration assistance? Contact training@upskilldevelopment.com for training dates, invoices and invitation letters.

Nairobi, Kenya Classroom/On-site Training Schedule

Course Date Location Fee Enroll
21/12/2026 to 01/01/2027 Nairobi 2,900 USD Register
18/01/2027 to 29/01/2027 Nairobi 2,900 USD Register
15/02/2027 to 26/02/2027 Nairobi 2,900 USD Register
15/03/2027 to 26/03/2027 Nairobi 2,900 USD Register
19/04/2027 to 30/04/2027 Nairobi 2,900 USD Register
17/05/2027 to 28/05/2027 Nairobi 2,900 USD Register
21/06/2027 to 02/07/2027 Nairobi 2,900 USD Register
19/07/2027 to 30/07/2027 Nairobi 2,900 USD Register
16/08/2027 to 27/08/2027 Nairobi 2,900 USD Register
20/09/2027 to 01/10/2027 Nairobi 2,900 USD Register
18/10/2027 to 29/10/2027 Nairobi 2,900 USD Register
15/11/2027 to 26/11/2027 Nairobi 2,900 USD Register
20/12/2027 to 31/12/2027 Nairobi 2,900 USD Register

Mombasa, Kenya Classroom/On-site Training Schedule

Course Date Location Fee Enroll
07/12/2026 to 18/12/2026 Mombasa 3,400 USD Register
18/01/2027 to 29/01/2027 Mombasa 3,400 USD Register
15/02/2027 to 26/02/2027 Mombasa 3,400 USD Register
15/03/2027 to 26/03/2027 Mombasa 3,400 USD Register
19/04/2027 to 30/04/2027 Mombasa 3,400 USD Register
17/05/2027 to 28/05/2027 Mombasa 3,400 USD Register
21/06/2027 to 02/07/2027 Mombasa 3,400 USD Register
19/07/2027 to 30/07/2027 Mombasa 3,400 USD Register
16/08/2027 to 27/08/2027 Mombasa 3,400 USD Register
20/09/2027 to 01/10/2027 Mombasa 3,400 USD Register
18/10/2027 to 29/10/2027 Mombasa 3,400 USD Register
15/11/2027 to 26/11/2027 Mombasa 3,400 USD Register
20/12/2027 to 31/12/2027 Mombasa 3,400 USD Register

Kigali, Rwanda Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Dubai, UAE Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Dar es Salaam, Tanzania Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Kampala, Uganda Classroom/On-site Training Schedule

Course Date Location Fee Enroll

Course Introduction

Generative artificial intelligence is reshaping how evaluation institutions plan studies, review documents, synthesise research, process qualitative data, prepare findings, and communicate evidence to decision-makers. These opportunities also introduce risks involving fabricated information, hidden bias, disclosure of sensitive information, intellectual property, inconsistent analytical quality, and overreliance on automated outputs. This course equips participants to establish governance arrangements that allow responsible innovation while protecting the independence, credibility, and public value of evaluation work.

The course focuses on the institutional policies, decision-making structures, controls, and assurance processes required to govern generative AI throughout its lifecycle. Participants examine how governance responsibilities should be allocated among executive leadership, evaluation managers, information technology teams, data protection officers, research specialists, procurement personnel, and independent assurance functions. They learn to translate broad principles such as transparency, accountability, fairness, privacy, security, and human oversight into operational procedures, approval gates, documented responsibilities, and measurable performance indicators.

A central component is practical AI risk management for evaluation-specific applications. Participants learn to identify intended uses, assess the sensitivity and reliability requirements of evaluation evidence, classify risks according to potential harm, and select proportionate safeguards. Case exercises address AI-assisted literature reviews, evidence synthesis, interview transcription, beneficiary feedback analysis, report drafting, and knowledge management. Participants practise identifying hallucinations, bias, unsupported conclusions, privacy exposures, automation bias, and other failure modes that could distort findings or undermine stakeholder confidence.

The programme also addresses data governance, cybersecurity, procurement, third-party model dependencies, and the handling of confidential or personally identifiable information. Participants explore how data should be classified before entering AI systems, how approved tools and vendors should be assessed, and how retention, access, logging, and deletion requirements can be documented. The course emphasises that contractual assurances and vendor claims do not replace institutional due diligence, and that controls must reflect the actual information being processed, the intended use, the consequences of error, and the applicable legal and ethical obligations.

Participants examine how to design an assurance framework that makes AI-assisted evaluation work auditable and reproducible. This includes maintaining records of tool versions, prompts, source materials, human reviews, methodological decisions, and changes made to AI-generated outputs. They learn to establish validation thresholds, escalation procedures, incident registers, review schedules, and monitoring indicators. The course also considers how to communicate AI involvement transparently without exaggerating system capabilities, concealing material limitations, or weakening the distinction between machine-generated suggestions and verified evaluation judgements.

By the end of the ten-day programme, participants will be prepared to develop an actionable governance roadmap for their institutions, connecting policy commitments with day-to-day evaluation workflows and management oversight. They will work toward practical outputs such as an AI-use policy, institutional risk register, approval matrix, vendor assessment checklist, human-review protocol, and monitoring dashboard. These deliverables are designed to support consistent practice across teams, improve organisational readiness, strengthen trust in evaluation evidence, and enable institutions to adopt generative AI in a controlled, proportionate, and continuously improving manner.

Duration

10 days

Who Should Attend

  • Directors and executive leaders responsible for institutional governance, strategic planning, organisational accountability, and responsible AI adoption.

  • Heads of monitoring, evaluation, accountability, and learning functions seeking reliable controls for AI-assisted evaluation workflows.

  • Evaluation managers responsible for study design, evidence quality, methodological integrity, reporting standards, and approval of analytical outputs.

  • Research directors and principal investigators overseeing mixed-methods studies, programme evaluations, policy research, and evidence synthesis.

  • Data governance officers responsible for information classification, access permissions, retention schedules, and responsible data use.

  • Data protection and privacy officers advising on personal information, confidentiality, consent, lawful processing, and privacy risk.

  • Information technology and cybersecurity managers responsible for AI platforms, access controls, system security, and vendor integrations.

  • Legal, compliance, and ethics professionals assessing contractual obligations, regulatory exposure, research ethics, and institutional liability.

  • Procurement and contract management specialists selecting AI vendors, negotiating safeguards, and monitoring supplier performance.

  • Quality assurance and internal audit professionals assessing control effectiveness, documentation, compliance, and evidence traceability.

  • Evaluation consultants and technical advisers supporting institutions with AI-enabled research, reporting, and knowledge management.

  • Programme managers and donor accountability specialists seeking to ensure that AI-supported findings remain credible and decision-useful.

  • Qualitative researchers and evidence synthesis specialists processing interviews, focus groups, case studies, and documentary evidence.

  • Knowledge management and learning officers responsible for institutional repositories, report libraries, and evidence-sharing practices.

  • Digital transformation and innovation leads designing responsible adoption pathways, staff guidance, and AI capability-building initiatives.

Course Objectives

  • Develop an institution-wide generative AI governance framework that connects strategic objectives, ethical principles, operational responsibilities, risk tolerance, and evaluation quality requirements.

  • Identify and classify generative AI use cases according to the sensitivity of information processed, the likelihood of failure, the severity of possible harm, and the consequences for evaluation decisions.

  • Establish clear accountability arrangements covering executive sponsors, evaluation leads, technical teams, data protection officers, procurement personnel, users, and independent assurance functions.

  • Apply recognised AI risk management concepts to evaluation workflows, translating broad governance principles into practical controls, approval processes, monitoring activities, and documented decisions.

  • Design data governance procedures that address confidentiality, personal information, informed consent, data minimisation, retention, access, secure transfer, and responsible disposal of evaluation materials.

  • Assess generative AI outputs for hallucinations, factual inaccuracies, representational bias, source misattribution, unsupported causal claims, and distortions of qualitative or quantitative evidence.

  • Establish human oversight protocols that specify when expert review is mandatory, how reviewers verify outputs, and when automated assistance must be restricted or escalated.

  • Develop vendor due diligence and procurement criteria addressing security, model transparency, data handling, contractual commitments, service continuity, change management, and exit arrangements.

  • Create auditable documentation standards for prompts, source records, model versions, validation results, reviewer decisions, corrections, and the final use of AI-assisted material.

  • Design incident management procedures for data exposure, misleading outputs, discriminatory results, security weaknesses, unauthorised tool use, and other failures affecting evaluation integrity.

  • Define performance indicators and assurance tests for governance maturity, control effectiveness, output reliability, staff compliance, user confidence, and the continuing suitability of approved AI applications.

  • Produce a practical implementation roadmap that prioritises high-value, proportionately controlled use cases and establishes ownership, timelines, resources, review milestones, and continuous improvement mechanisms.

Comprehensive Course Outline

Module 1: Foundations of Generative AI Governance in Evaluation Institutions

  • Distinguish generative AI, conventional automation, predictive analytics, and retrieval-based systems, focusing on their different capabilities, limitations, and governance implications for evaluation practice.

  • Examine institutional applications such as literature synthesis, qualitative coding, survey analysis support, report drafting, evidence discovery, and knowledge management, identifying where human judgement remains essential.

  • Analyse how unreliable outputs, unclear accountability, and inappropriate data handling can affect evaluation independence, methodological integrity, stakeholder trust, and decision-making.

  • Map the relationship between organisational strategy, research ethics, evaluation standards, information governance, and responsible AI adoption to establish a coherent institutional approach.

Module 2: Governance Principles, Leadership and Institutional Accountability

  • Define governance principles covering transparency, fairness, reliability, privacy, security, human oversight, explainability, proportionality, and accountability in AI-supported evaluation activities.

  • Allocate decision rights and responsibilities among governing boards, executive management, evaluation leadership, IT teams, legal advisers, data stewards, and frontline users.

  • Design an AI governance committee or equivalent oversight mechanism with clear terms of reference, decision authority, escalation pathways, reporting obligations, and meeting schedules.

  • Develop policy ownership, approval thresholds, exception procedures, and accountability records that prevent responsibility from becoming fragmented across institutional functions.

Module 3: Generative AI Risk Identification and Classification

  • Build an inventory of AI applications, users, vendors, datasets, workflows, integrations, and intended outputs to establish a reliable institutional picture of AI exposure.

  • Assess inherent risk by considering the likelihood of failure, sensitivity of input data, scale of use, affected populations, reversibility of decisions, and consequences of incorrect outputs.

  • Classify evaluation use cases into risk tiers and define proportionate approval requirements, restrictions, validation standards, human review, and monitoring intensity for each tier.

  • Maintain a living risk register that records causes, impacts, existing controls, residual risk, accountable owners, mitigation actions, deadlines, and review triggers.

Module 4: Legal, Regulatory, Ethical and Standards Alignment

  • Compare relevant data protection, intellectual property, confidentiality, public-sector, research ethics, and AI-specific requirements across the jurisdictions in which an institution operates.

  • Examine the NIST AI Risk Management Framework, its Generative AI Profile, and ISO/IEC 42001 as complementary resources for risk management and institutional management-system design.

  • Assess how the EU AI Act may apply to an institution's activities, suppliers, users, and affected individuals, taking account of territorial scope and the applicable implementation timetable.

  • Establish a compliance mapping process that links each relevant obligation to an internal policy, accountable role, evidence requirement, monitoring control, and corrective action procedure.

Module 5: Data Governance, Privacy and Confidentiality Controls

  • Create data classification rules for public information, internal working documents, confidential evaluation records, personal data, sensitive beneficiary information, and restricted research materials.

  • Apply data minimisation, purpose limitation, appropriate access controls, secure storage, retention limits, and deletion procedures before information enters a generative AI workflow.

  • Evaluate privacy risks associated with prompts, uploaded documents, conversation histories, model training practices, logging, cross-border transfers, and external service integrations.

  • Develop procedures for consent and disclosure where applicable, privacy impact assessment, anonymisation or pseudonymisation, secure processing, and review of third-party data handling commitments.

Module 6: AI Model Selection, Procurement and Vendor Due Diligence

  • Define functional, methodological, security, privacy, accessibility, interoperability, and cost criteria for selecting generative AI tools that meet institutional evaluation requirements.

  • Examine vendor claims about accuracy, confidentiality, data retention, training use, model updates, service availability, and security certifications, identifying evidence needed to substantiate them.

  • Design procurement questionnaires and scoring matrices that assess supplier transparency, subcontractors, incident notification, audit rights, data portability, service termination, and contractual accountability.

  • Establish controlled pilot testing, vendor approval, periodic reassessment, and exit planning so that supplier changes or discontinued services do not undermine evidence continuity.

Module 7: Evaluation Integrity, Reliability and Output Validation

  • Identify hallucinations, fabricated citations, misleading summaries, incorrect calculations, incomplete evidence, and overconfident explanations in AI-generated evaluation materials.

  • Develop verification protocols requiring traceable sources, corroboration against original records, independent checks of numerical claims, and documented resolution of material discrepancies.

  • Set different acceptance thresholds for low-risk administrative support, research synthesis, analytical interpretation, public reporting, and outputs influencing consequential decisions.

  • Design sampling and review procedures to assess output accuracy, consistency, completeness, and reproducibility while retaining human responsibility for methodological conclusions.

Module 8: Bias, Fairness, Representation and Inclusion

  • Identify potential sources of bias in training data, source selection, language coverage, prompt formulation, classification categories, and model-generated interpretations of stakeholder evidence.

  • Assess how generative AI may underrepresent minority voices, misinterpret local expressions, reproduce stereotypes, or distort the experiences of groups facing structural disadvantage.

  • Design disaggregated review procedures, community-informed validation, multilingual checks, and appropriate fairness indicators for evaluation contexts and affected populations.

  • Document limitations and corrective actions where the evidence base is incomplete, representational harms cannot be fully mitigated, or the model is unsuitable for a proposed application.

Module 9: Human Oversight and Decision Rights

  • Define meaningful human oversight by specifying reviewer competence, access to original evidence, authority to reject outputs, time for review, and accountability for final decisions.

  • Create approval gates for high-impact outputs, sensitive-data processing, external publication, automated classifications, and interpretations that may influence programme or funding decisions.

  • Address automation bias and overreliance by training reviewers to challenge plausible but unsupported outputs and to preserve independent analytical reasoning.

  • Establish escalation and override procedures that allow staff to pause AI use, seek specialist review, document disagreements, and report concerns without undue pressure.

Module 10: Cybersecurity, Prompt Injection and Technical Risk

  • Examine threats including prompt injection, malicious documents, data leakage, insecure connectors, compromised credentials, unsafe plug-ins, and unintended access to institutional repositories.

  • Apply least-privilege access, identity management, secure configuration, approved integrations, logging, secrets management, and separation of sensitive workloads.

  • Assess how retrieved documents and external content can contain untrusted instructions that manipulate model behaviour or expose information through connected workflows.

  • Establish technical testing, vulnerability reporting, incident escalation, recovery arrangements, and periodic security reviews proportionate to the institution's threat environment.

Module 11: Intellectual Property, Attribution and Research Provenance

  • Evaluate ownership, licensing, reuse permissions, confidentiality restrictions, and attribution requirements for source materials, AI-generated text, datasets, and commissioned research products.

  • Establish documentation practices that preserve the origin of evidence, distinguish quotations from paraphrases, and prevent fabricated or misrepresented references from entering evaluation reports.

  • Define when AI assistance should be disclosed in research protocols, deliverables, publication notes, internal records, or communications with commissioning organisations.

  • Develop review procedures for third-party content, copyrighted materials, confidential submissions, and outputs that may reproduce protected or restricted source information.

Module 12: AI Governance for Evaluation Workflows and Evidence Pipelines

  • Map where generative AI enters the evaluation lifecycle, from question formulation and desk review to data collection support, analysis, interpretation, reporting, and organisational learning.

  • Define permitted and prohibited uses at each workflow stage, including requirements for source preservation, methodological review, human approval, and traceability of AI-assisted changes.

  • Integrate governance checkpoints into project initiation, research ethics review, data management planning, analysis protocols, report clearance, and final evidence archiving.

  • Design workflow documentation that connects source records, prompts, model outputs, reviewer interventions, methodological decisions, and approved deliverables in a reproducible chain.

Module 13: Third-Party Risk, Partnerships and Institutional Collaboration

  • Assess governance risks arising when consultants, research partners, implementing organisations, donors, or subcontractors use generative AI in shared evaluation activities.

  • Establish contractual clauses and partner protocols for approved tools, confidentiality, data access, incident reporting, evidence validation, auditability, and responsible disclosure.

  • Develop collaborative review arrangements that define who verifies AI-assisted outputs, resolves disagreements, manages cross-border information, and approves shared findings.

  • Monitor third-party compliance through risk-based assurance checks, evidence sampling, corrective action plans, and documented review of material supplier or partner changes.

Module 14: Incident Response, Complaints and Corrective Action

  • Define reportable AI incidents, including confidential data exposure, fabricated evidence in published reports, harmful bias, unauthorised use, security compromise, and material analytical errors.

  • Establish reporting channels, severity classifications, notification responsibilities, containment actions, evidence preservation, investigation procedures, and communication protocols.

  • Conduct root-cause analysis to determine whether failures arose from policy gaps, unsuitable models, poor data quality, insufficient review, configuration weaknesses, or supplier practices.

  • Track remediation, affected-output correction, stakeholder notification where required, lessons learned, and approval to resume use after the relevant risks have been addressed.

Module 15: Assurance, Monitoring and Governance Performance

  • Develop a governance assurance plan with measurable indicators for policy compliance, risk treatment, validation quality, reviewer participation, incident trends, and corrective action closure.

  • Design periodic audits and sample-based reviews to test whether actual AI use matches approved purposes, data restrictions, validation procedures, and documentation requirements.

  • Monitor model changes, emerging threats, shifting vendor terms, user behaviour, stakeholder complaints, and new regulatory guidance that may affect previously approved applications.

  • Use assurance findings to update risk classifications, strengthen controls, improve staff capability, retire unsuitable tools, and report institutional AI risk to decision-makers.

Module 16: Governance Roadmap, Simulation and Institutional Action Plan

  • Conduct a baseline assessment of governance maturity, existing policies, technical safeguards, staff skills, procurement arrangements, and current generative AI use cases.

  • Prioritise improvements according to risk exposure, operational value, available resources, implementation dependencies, and the urgency of unresolved governance gaps.

  • Complete a practical simulation involving an AI-assisted evaluation workflow, identifying risks, applying approval controls, validating outputs, documenting decisions, and responding to an incident.

  • Present an implementation roadmap with named owners, milestones, resource requirements, performance indicators, reporting arrangements, and scheduled reviews for continuous improvement.

Training Approach

This course will be delivered by our skilled trainers who have vast knowledge and experience as expert professionals in the fields. The course is taught in English and through a mix of theory, practical activities, group discussion and case studies. Course manuals and additional training materials will be provided to the participants upon completion of the training.

Tailor-Made Course

This course can also be tailor-made to meet organization requirement. For further inquiries, please contact us on: Email: training@upskilldevelopment.com Tel: +254 721 331 808

Training Venue 

The training will be held at our Upskill Training Centre. We also offer training for a group (at a discount of 10% to 50%) at requested location all over the world. The Onsite course fee covers the course tuition, training materials, two break refreshments, buffet lunch, airport transfers, Upskill gift package, and guided tour.

Visa application, travel expenses, dinners, accommodation, insurance, and other personal expenses are catered by the participant

Certification

Participants will be issued with Upskill certificate upon completion of this course.

Airport Pickup and Accommodation

Airport pickup and accommodation is arranged upon request. For booking contact our Training Coordinator through Email: training@upskilldevelopment.com, +254 721 331 808

Terms of Payment:

Unless otherwise agreed between the two parties’ payment of the course fee should be done 3 working days before commencement of the training so as to enable us to prepare better.

Choose Your Training Format

Attend a scheduled session or request a customized programme for your organization.

Classroom Training

Attend instructor-led training at one of our scheduled training locations.

Live Online Training

Join interactive instructor-led sessions remotely from your office or home.

In-House Training

Customize the course content, dates and delivery for your team or organization.

Certification

Participants who successfully complete the programme receive a Certificate of Completion from Upskill Development Institute.

Request Course Information

Need a Tailor-Made Programme?

We can customize the course content, duration, schedule and delivery format to match your organization's objectives.

Dates & Fees Enquire Now

Organizations we've served

Some of Our Recent Clients

We support professional development and institutional capacity building for organizations across a wide range of sectors. Explore a selection of organizations featured in our client portfolio.

SNV logoSNV
National Communications Authority logoNational Communications Authority
Masinde Muliro University of Science and Technology logoMasinde Muliro University of Science and Technology
Kenya Medical Research Institute logoKenya Medical Research Institute
Save Lamu logoSave Lamu
UNICEF logoUNICEF
Fairtrade Africa logoFairtrade Africa
World Agroforestry (ICRAF) logoWorld Agroforestry (ICRAF)
CPI logoCPI
Food and Agriculture Organization (FAO) logoFood and Agriculture Organization (FAO)
University of Southampton logoUniversity of Southampton
Ngali Holdings logoNgali Holdings
STOPAIDS logoSTOPAIDS
International Fertilizer Development Center (IFDC) logoInternational Fertilizer Development Center (IFDC)
Wide Communications logoWide Communications

Learn. Apply. Grow.

Build skills that make a difference at work

Upskill Development Institute offers practical professional training, customized group programmes and flexible learning options to help individuals and organizations strengthen performance.

Chat with us on WhatsApp