Oops! Sorry!!


This site doesn't support Internet Explorer. Please use a modern browser like Chrome, Firefox or Edge.

AB-731 Revision App | BFT Consulting
BFT

AB-731 Revision App

AI Transformation Leader Β· Full Study Pack

EXAM READY

Azure AI Foundry Services

8 core services β€” tap each card to expand

Memory Key
COST of Very Large ValueS
Content Safety Β· OpenAI Β· Speech Β· Translator Β· Vision Β· Language Β· Video Indexer Β· Search
πŸ›‘

AI Content Safety

The Digital Bouncer

Scans text and images to detect and block hate speech, violence, sexual content, or self-harm material before anyone sees it.

Nigerian Use Case

A Lagos online gaming platform blocks offensive messages in real time before they reach other players.

πŸ€–

Azure OpenAI Service

The Generative Brain

Enterprise-grade access to GPT, DALLΒ·E, and embeddings inside Azure's secure infrastructure β€” for chat, summarisation, and code generation.

Nigerian Use Case

GTBank builds an internal assistant that helps analysts summarise credit reports and draft client communications.

πŸŽ™

AI Speech

The Voice Converter

Converts voice to text and text to natural-sounding voice. Also handles real-time translation, voice cloning, and speaker recognition.

Exam Pipeline

AI Speech + AI Language = speak notes β†’ transcribe β†’ detect mood. Classic two-service combo.

Nigerian Use Case

A Nigerian insurance company transcribes call centre recordings automatically β€” cutting note-taking time by 70%.

🌐

AI Translator

The Polyglot

Instant text and full document translation across dozens of languages β€” preserving original formatting. NOT the same as AI Language.

Nigerian Use Case

Dangote Cement translates English contracts into French and Portuguese for West African partners in seconds.

πŸ‘

AI Vision

The Digital Eye

Extracts insights from still images β€” object detection, OCR, image captioning, facial recognition. Evolved from Form Recognizer.

Nigerian Use Case

Nigerian Ports Authority scans shipping container labels automatically instead of manual data entry.

πŸ’¬

AI Language

The Text Analyst β€” 4 Pre-Built Tools

🎭 Sentiment Analysis β€” detects positive/neutral/negative mood
🏷 Entity Recognition β€” spots names, dates, places, companies
πŸ”‘ Key Phrase Extraction β€” summarises main points
🌐 Translation β€” converts text between languages

Exam Trap

Do NOT build a custom classification model when AI Language already has Sentiment Analysis out of the box.

Nigerian Use Case

Stanbic IBTC flags negative customer feedback from thousands of reviews automatically.

🎬

AI Video Indexer

The Video Reviewer

Analyses video to extract transcriptions, detect faces, segment scenes, identify topics β€” creating searchable video archives.

Nigerian Use Case

Channels TV searches its archive to find all clips where a specific politician spoke, without watching hours of footage.

πŸ”

AI Search

The Smart Librarian

Searches across your organisation's private internal documents using semantic, vector, and hybrid search.

Critical Trap

AI Search β‰  File Search in Copilot Studio. AI Search = Azure enterprise service. File Search = Copilot Studio uploaded files.

Nigerian Use Case

Templars Law Firm finds exact contract clauses from plain-language questions.

⚑ Entity Recognition vs Sentiment AnalysisHighlight names, dates, places β†’ Entity Recognition. Detect if customer is angry/happy/neutral β†’ Sentiment Analysis. Both live inside Azure AI Language.

Rent, Buy, Build / Own

The AI ecosystem sourcing strategy β€” core AB-731 exam domain

The Three Strategies at a Glance

StrategyWhat It MeansMicrosoft ExampleWhen to Use
πŸͺ Rent (SaaS)Subscribe to ready-to-use AI apps. Zero coding or infrastructure needed.Microsoft 365 CopilotStaff need AI in existing M365 apps NOW. No IT budget for development.
πŸ›’ Buy (API/PaaS)Purchase AI APIs or platforms and integrate them into your own systems.Azure OpenAI ServiceNeed custom interface or logic, but no in-house ML team to train models.
πŸ— Build/OwnDevelop bespoke AI models from scratch using your own data science team and proprietary data.Training a custom LLMHighly regulated data (NDPR), unique competitive advantage, very large scale.

The Exam Golden Rule

πŸš• Rent = Taking a Taxi

You say "Take me to the airport." You don't touch the steering wheel or know the route. Microsoft's AI does all the complex work. You just configure what you want.

πŸš— Build = Renting a Car

You must build the route, steer, press the pedals yourself. Developers must write code, tune the AI, and test it before anyone can use it. Takes time and skills you may not have.

Decision Triggers β€” What to Look for in Exam Questions

If the scenario says...StrategyReasoning
"No available IT budget for programming" / "zero coding"πŸͺ RentOnly SaaS works without development effort
"Deploy quickly" / "out-of-the-box" / "just turn it on"πŸͺ RentSaaS is ready immediately
"Custom interface" but "no ML team"πŸ›’ BuyAPI/PaaS gives custom interface without training models
"Strict NDPR / data residency" / "proprietary data"πŸ— Build/OwnSensitive data can't leave your environment
"Unique competitive advantage" / "core IP"πŸ— Build/OwnJustifies the cost of training from scratch
"API cost at scale getting too high" (volume trap)πŸ— Build/OwnAt sufficient volume, building own model beats API bills
"Extending Microsoft 365 Copilot with custom agents"πŸ— Build"Custom" = development required = violates "no coding" requirement

Nigerian Business Scenarios

OrganisationScenarioStrategy
FMCG HR ChatbotLagos FMCG company needs HR chatbot inside Teams, no coding budgetπŸͺ Rent (M365 Copilot)
Fintech KYCFintech needs custom KYC verification UI with API integration, no ML teamπŸ›’ Buy (Azure OpenAI API)
NGX Trading FirmTrading firm wants proprietary market prediction using 10 years of internal data β€” core competitive IPπŸ— Build/Own
HealthTech / NDPRHospital cannot send patient data to Microsoft cloud due to NDPR complianceπŸ— Build/Own
Japa TurnoverBuilt ML team, key engineers left to Europe ("Japa syndrome") β€” pivot neededπŸ›’ Buy (API/PaaS)
⚑ The Rent vs Adopt TrapWhen a scenario requires a "quick, zero-coding solution", "Adopt Microsoft 365 Copilot Chat" = Rent. "Extend with custom agents" = Build. Even though "extending" sounds smaller, the word "custom" means development effort is required β€” which violates any "no coding" constraint.
⚑ Declarative Agent β€” The Taxi ApproachA "declarative agent" means you describe what you want without writing code β€” like telling the taxi driver your destination. You configure existing Copilot with specific instructions and data sources. Look for "no coding" or "no development" in the scenario β€” that points to a declarative agent.

Embedded vs Stand-Alone Agents

Align AI architecture to the right business problem

Side-by-Side Comparison

Feature🧩 Embedded Agent🏒 Stand-Alone Agent
DefinitionAI integrated directly into an existing application or workflowIndependent AI in its own dedicated platform or portal
User InterfaceLives inside software users already use (sidebar, chat bubble, inline)Has its own distinct web page, app, or portal
Context AwarenessAutomatically knows what you are working on in that appBlank slate β€” you must provide full context every time
Primary BenefitFrictionless β€” no context switching neededHighly flexible; handles broad, complex, cross-domain tasks
ExamplesMicrosoft 365 Copilot inside Word, Teams, OutlookChatGPT, Claude, a custom portal built on Copilot Studio
Build/Buy LinkBuy or Extend (purchase M365 Copilot)Build (Copilot Studio or Azure AI)

The Office Building Analogy

🧩 Embedded = Colleague at Your Desk

You're designing a proposal in Word. A colleague stands right next to you looking at the same screen. You say "make this punchier" β€” they fix it instantly. Zero briefing needed. Full context.

🏒 Stand-Alone = Help Desk Down the Hall

You must stop work, walk to a different office, and fully brief the expert: "I'm building a training program, here's my outline..." They give brilliant answers but you carry the result back yourself.

⚑ Why This is on the AB-731 Exam β€” 35–40% of the TestThe exam maps business processes to the right Microsoft AI tool. Embedded = frictionless, contextual, existing M365 workflow. Stand-Alone = complex, specialised, custom data outside M365. Embedded agents also reduce adoption friction in change management.

M365 Copilot vs Copilot Studio

Microsoft 365 CopilotMicrosoft Copilot Studio
What it isReady-to-drive car. Already built. Lives inside Teams, Word, Outlook.Mechanic's garage. You build your own custom bot or connection here.
When to useSummarise Teams meetings, read M365 emails, draft Word docs β€” out of the boxConnect to 3rd-party systems (Workday, Salesforce), build custom chatbots
Development needed?No β€” just assign licenses and turn it onYes β€” configuration, logic building, and integration work required
Exam trigger word"Out-of-the-box", "within M365", "no development""Custom", "3rd-party integration", "build a bot", "Workday/Salesforce"
⚑ The "Developing a Solution" TrapIf the exam says "developing an AI solution to summarise Teams meetings and read Outlook emails" β€” the word "developing" makes you think Copilot Studio. WRONG. If the tasks are all native M365 activities (Teams, Outlook, Word), you need Microsoft 365 Copilot β€” it already does all of that out of the box.

Security, Microsoft Graph & RBAC

How data is protected when Copilot answers your questions

How Microsoft Graph Controls What Copilot Can See

1. You ask Copilot a question

"Summarise project updates from my team." Copilot does not own or directly store your company's data.

2. Copilot asks Microsoft Graph for the data

Copilot cannot just wander around opening locked cabinets. It must go through Microsoft Graph β€” the Head Librarian and Security Guard.

πŸ›‘

3. Microsoft Graph checks your badge (RBAC)

Graph does two things: (a) finds connected documents relevant to your query, and (b) checks your security clearance β€” it will ONLY hand over files you personally have permission to read.

πŸ“

4. Only allowed data reaches Copilot

If HR salary files are restricted, Graph blocks them. Copilot will say "I couldn't find that information" or "you don't have access."

The Corporate Library Analogy

πŸ“š Microsoft Entra ID = The Badge Office

Issues you a digital ID badge based on your job title. "This person is an HR Manager." It creates your Role-Based Access Control (RBAC) identity. Without this badge, you have no permissions at all.

πŸ” Microsoft Graph = The Security Librarian

Checks your badge before giving any file to Copilot. If you ask about the CEO's salary but you're not in the payroll group β€” Graph blocks it. The AI did not fail; the permissions allowed it.

RBAC, Monitoring, Purview β€” Active vs Passive

Control TypeWhat It DoesAnalogyEnough on its own?
Centralized MonitoringRecords what AI does β€” logs, audit trails, dashboardsSecurity camera over the vault β€” records the robbery but doesn't stop it❌ No β€” passive only
RBAC + Controlled InterventionRestricts what AI can DO β€” active enforcement, halts unsafe actions in real timeCombination lock + automatic steel doors that drop the moment an alarm triggersβœ… Yes β€” active control
Microsoft Purview / DLPLabels documents and blocks Copilot from sharing sensitive contentDigital stamps on every file + scanner at the door checking each oneβœ… Yes β€” prevents oversharing
⚑ The Monitoring TrapWhenever a scenario asks to STOP, PREVENT, or INTERVENE against unsafe AI behaviour β€” monitoring is NEVER the complete answer. You need RBAC + controlled intervention to physically stop the AI from crossing boundaries.

Tenant Boundaries

ConceptWhat It MeansCorrect?
βœ… Isolation within tenant boundaryYour company's data stays locked on your floor of Microsoft's skyscraper. Other companies cannot see it.βœ… Correct enterprise benefit
❌ Public sharing across tenantsYour AI outputs shared with every other company❌ Catastrophic data leak
❌ Training models on shared customer dataMicrosoft blends your private files into its public AI training❌ Your strategy leaks to competitors
❌ Unrestricted anonymous accessAnyone walks in and uses your AI tools without a badge❌ Violates every security rule

Microsoft AI Licensing

M365 vs Azure β€” the two sides of the house

Golden Rule
Employees using M365 apps = Per-user license
Developers building custom Azure workloads = Azure billing
These are two completely different billing worlds β€” never mix them up

The Three Licensing Models

ModelHow It's BilledAnalogyBest Used When...
🎟 Copilot Add-on LicenseFixed price per user, per monthUnlimited gym membership β€” same cost whether you go once or every dayEmployees use Copilot daily in Word/Teams/Outlook. Predictable monthly cost.
⚑ Pay-As-You-Go (Azure)Billed on actual usage β€” like an electricity billElectric utility bill β€” you only pay for what you actually use each monthIrregular, seasonal usage. Custom Azure AI workloads. Field agents who use AI only 1–2x a month.
🏭 Prepaid AI Capacity CommitmentUpfront cost for reserved raw computing power in AzureRenting a dedicated private server room β€” you own it 24/7 regardless of useDevelopers running large-scale production workloads with stable, predictable high-volume usage on Azure.

The Two Sides of the Microsoft House

🏒 Side 1: Microsoft 365 (SaaS)

For employees using standard Office apps.
Scenario clues: "Give HR staff access to Copilot in Word/Teams", "assign Copilot seats", "out-of-the-box"
β†’ Copilot per-user add-on license

☁️ Side 2: Azure Cloud (PaaS/IaaS)

For developers building custom AI on Azure.
Scenario clues: "deploy custom Azure AI service", "production workload", "large-scale stable usage"
β†’ Pay-As-You-Go or Prepaid Capacity Commitment

⚑ The Prepaid Capacity Trap"Prepaid AI Capacity Commitment" exists β€” but ONLY for Azure production workloads with stable, high-volume usage. It does NOT apply to Microsoft 365 Copilot seat licensing for employees. When the scenario says "deploying Copilot features to HR staff in Word/Teams" β†’ per-user add-on. When it says "large-scale production Azure workload with predictable stable usage" β†’ prepaid capacity.

Partner Models in Azure AI Foundry β€” The Marketplace Trap

ConceptExplanation
What are Partner Models?3rd-party AI models (e.g. Meta, Mistral) hosted inside Azure AI Foundry's "cafeteria" β€” not Microsoft's own models
Billing methodAlways billed through the Azure Marketplace β€” NOT through your standard Azure subscription by default
Why it failsYour standard Azure subscription may be locked to direct Microsoft purchases only. Trying to use a partner model gets "declined."
The fixEnable Marketplace purchases on your existing Azure subscription β€” not creating a new subscription, not buying Copilot Studio packs
⚑ Licensing Summary Cheatβ€’ 200 full-time HR staff using Copilot daily in Word/Teams β†’ Copilot per-user add-on license
β€’ 200 field agents using AI only 1–2x per month β†’ Pay-As-You-Go (Azure)
β€’ Custom AI chatbot on Azure website with unpredictable traffic β†’ Pay-As-You-Go (Azure)
β€’ Large-scale Azure production AI workload, stable usage, finance wants flat bill β†’ Prepaid Capacity Commitment

LLM vs SLM Β· Fine-tuning vs Pretrained Β· RAG

Model architecture decisions β€” critical for AB-731

LLM vs SLM β€” The Transport Analogy

FeatureπŸš› LLMπŸ›΅ SLM
Brain PowerDeep reasoning, multi-step logic, complex analysisSpecific, straightforward, focused tasks
SpeedSlower β€” takes more compute timeVery fast β€” low latency
CostHigher cost per requestMuch lower cost per request
Where it runsRequires powerful cloud serversCan run locally on phones and laptops without internet
The AnalogySemi-truck β€” hauls massive complex cargo, needs a lot of fuelDelivery scooter β€” fast, cheap on fuel, runs anywhere
Microsoft ExampleGPT-4o via Azure OpenAIMicrosoft Phi series (Phi-3, Phi-4)
⚑ SLM Trigger Words"Lightweight task" / "low latency" / "edge deployment" / "on-device / offline" / "cost-constrained" / "runs on mobile without internet" β†’ Always SLM
"Complex reasoning" / "multi-step logic" / "synthesise multiple data sources" / "deep analysis of unstructured data" β†’ Always LLM

Pretrained vs Fine-Tuning vs RAG

ApproachWhat It MeansAnalogyWhen to Use
πŸ“š Pretrained ModelGeneral AI trained on massive internet data. Knows many things but not specialised.New graduate with a general degree β€” smart, can talk about anything, not a legal expert yetGeneral tasks. Good starting point.
πŸŽ“ Fine-TuningExtra training on a specific, smaller dataset. Mathematically adjusts internal "weights" β€” it becomes a specialist.Sending the general graduate to law school β€” focused training until they master that domainWhen AI must use specialised terminology reliably (legal, medical, finance).
πŸ“‘ RAGRetrieves external data at query time and feeds it as context. Internal weights are NOT changed.Giving a smart generalist a reference booklet to read before answering each questionWhen you need the model to know your private documents without modifying the model.
⚑ The Critical Distinction β€” Internal WeightsFine-tuning MODIFIES internal weights = model's brain is permanently changed for that domain.
RAG does NOT modify internal weights = model stays the same, just reads new documents at query time.
Key phrase: "without modifying internal weights" β†’ RAG. "Model must reliably use specialised terminology" β†’ Fine-tuning.

Copilot Tools

Pages, Notebooks, Dynamics 365 BC, and how they compare

πŸ“„

Copilot Pages

Real-time collaborative canvas

Turn an AI chat response into an editable document your team can work on side-by-side in real time. Lightweight, fast content creation β€” proposals, brainstorms.

Equivalent

β‰ˆ Claude Artifact β€” the side-by-side editing window that opens when Claude generates a document or code

πŸ““

Copilot Notebooks

Structured project workspace

Bundle multiple files, emails, Pages, and references together. AI reasons across all of it for long, complex projects. Closest to Custom GPT β€” add custom instructions + upload knowledge files.

Equivalent

β‰ˆ Claude Projects AND β‰ˆ OpenAI Custom GPT

Copilot in Dynamics 365 Business Central

FeatureWhat It DoesNigerian Example
πŸ“ AI Item DescriptionsWrites product catalogue descriptions automaticallyLagos FMCG auto-generates descriptions for 5,000 SKUs in minutes
🏦 Bank Reconciliation AssistAI matches bank transactions to ledger entriesNigerian manufacturer reconciles accounts in minutes, not days
πŸ›’ Sales Order SuggestionsAnalyses sales history and suggests what to reorderKano distributor gets automatic reorder recommendations
πŸ’° Cash Flow ForecastingPredicts future cash position from receivables and payablesLagos trading company CFO sees 90-day cash forecast automatically
πŸ“’ Marketing Text SuggestionsGenerates promotional copy ready for SMS, email, or socialNigerian retail chain generates SMS content for new product launches instantly
⚑ Dynamics NAV vs Dynamics 365 BCMany Nigerian companies still run Dynamics NAV on-premise. To access ANY Copilot features above, they MUST migrate to Dynamics 365 Business Central (cloud). This is a strategic AI transformation advisory opportunity for BFT Consulting clients β€” and an exam trap.

πŸƒ Flashcards

Tap a card to flip β€” work through each set until you know them all

0 / 0 known
QUESTION β€” tap to flip
Loading...
πŸ‘† Tap anywhere on this card to reveal the answer
ANSWER
πŸ‘† Tap to flip back
Card 0 of 0

Practice Quiz

50 scenario-based questions β€” choose how many to attempt

How many questions would you like?

Minimum 2 questions per category guaranteed in every session.

Questions Correct

Exam-Day Cheat Sheet

Everything to hold in your head β€” one page

COST VLVS β€” 8 Azure AI Services
C β€” Content SafetyDetects harmful text/images (The Bouncer)
O β€” OpenAI ServiceGPT, DALLΒ·E, enterprise AI generation
S β€” SpeechVoice ↔ text, real-time (The Voice Converter)
T β€” TranslatorFull documents across languages
V β€” VisionReads images and OCR (The Digital Eye)
L β€” LanguageSentiment Β· Entity Β· Key Phrase Β· Translation
V β€” Video IndexerTranscribes + indexes video content
S β€” SearchSemantic search on your private data
Rent / Buy / Build Decision
Rent (SaaS)No coding, use M365 Copilot, fast deploy
Buy (API/PaaS)Custom UI, no ML team, Azure OpenAI
Build/OwnNDPR, proprietary IP, very high scale
Rent trigger"No budget", "out-of-the-box", "immediate"
Build trigger"Custom", "develop", "3rd-party integration"
Declarative agentConfigure only β€” no code, like taking a taxi
Licensing β€” Two Sides of the House
M365 Per-UserEmployees using standard M365 apps daily
Pay-As-You-GoCustom Azure workloads, irregular usage
Prepaid CapacityAzure production workload, stable high volume
NOT PrepaidNever for M365 Copilot seat licensing
Partner ModelsEnable Azure Marketplace on existing subscription
Critical Exam Traps
File SearchCopilot Studio feature (uploaded files)
AI SearchAzure service (semantic, private org data)
RAG No weight change β€” external data at query time
Fine-tuningModifies internal weights β€” specialist training
Monitoring aloneNever enough β€” passive, doesn't prevent harm
AI Language β‰  TranslatorLanguage = NLP toolbox, Translator = lang conv
Search Types
KeywordExact words, strict filter. Rigid.
Conceptual/VectorMeaning + synonyms. No filter.
HybridBoth: filter a category AND find meaning
Semantic RankingReorders existing results only. No searching.
Hybrid trigger"Related concepts" + "limit to category"
Security & Models
Entra IDIssues digital ID badges (RBAC roles)
Microsoft GraphSecurity librarian β€” checks badge before giving files
Purview DLPBlocks Copilot sharing sensitive data
Tenant boundaryYour data stays on your floor. Always correct answer.
LLMComplex reasoning, cloud, expensive, slower
SLMLightweight, offline, fast, cheap (Microsoft Phi)
Embedded vs Stand-Alone Agents
EmbeddedLives inside M365 apps, auto-context, no switch
Stand-AloneOwn portal, blank slate, you provide context
M365 CopilotOut-of-box embedded. Teams/Word/Outlook.
Copilot StudioBuilder for custom bots + 3rd-party connections
"Developing" trapIf tasks = native M365, answer is M365 Copilot
Responsible AI β€” FAITHP
F β€” FairnessNo bias or discrimination across groups
A β€” AccountabilityHumans responsible for AI decisions
I β€” InclusivenessBenefits all people, all abilities
T β€” TransparencyExplainable decisions β€” no black boxes
H β€” Human OversightHumans stay in control
P β€” PrivacyData protection by design