Explanation:
Ensure consistent AI responses # Define standardized prompt templates; Support governance and version control # Store prompts in a Git repository The requirement says the shared prompt library must ensure consistent AI responses with reusable formats.
The best way to achieve that is to create standardized prompt templates.
Standardized templates help multiple business units use the same structure for:
system instructions
response style
output formatting
safety guidance
business terminology
escalation behavior
This is important in AI-powered business solutions because prompt variability often leads to inconsistent outputs. A shared template approach creates repeatability and reduces drift across departments.
From an enterprise AI design perspective, standardized prompt templates improve:
response consistency
quality control
reusability
onboarding speed
compliance alignment
They also minimize administrative effort because teams do not need to invent new prompt structures from scratch.
Why "Store prompts in a Git repository" is correct
The requirement says the solution must support governance and version control while minimizing effort and cost.
A Git repository is the most appropriate answer because it provides:
version history
change tracking
rollback capability
collaboration support
approval workflows
auditability
This directly supports governance. It allows the organization to know:
who changed a prompt
what changed
when it changed
why it changed
That is extremely valuable for enterprise AI operations, especially when prompts influence business decisions, customer interactions, compliance behavior, or internal process automation.
Git also minimizes ongoing cost because it uses standard development and content management practices rather than requiring a separate expensive governance platform for prompt assets.
Why the other options are incorrect
Delegate department-specific prompt templates
This reduces consistency instead of improving it. Different departments creating their own templates introduces fragmentation, style drift, and governance challenges.
Maintain a prompt history
Keeping a history is useful, but it does not by itself ensure consistent reusable formats. It is more of a recordkeeping activity than a standardization mechanism.
Categorize prompts by business function
Categorization is helpful for organization and discoverability, but it does not provide version control or governance by itself.
Define standardized prompt templates for governance/version control
Standardization improves consistency, but version control specifically requires a system for tracking and managing changes. That is why Git is the better answer for the governance requirement.
Expert reasoning
A shared enterprise prompt library should usually follow this pattern:
Consistency and reuse # standard templates
Governance and versioning # Git-based repository
Low cost and low admin overhead # reuse existing tooling and centralized controls

Comprehensive and Detailed Explanation From Agentic AI Business Solutions Topics:
The correct answer is B. a horizon-based framework.
This question is about evaluating ROAI across a portfolio of AI initiatives that are at different stages of development. The key requirement is to use a structured approach that balances:
immediate results
long-term value
strategic innovation
That wording maps directly to a horizon-based framework.
Why B is correct
A horizon-based framework is designed to evaluate investments across different time horizons, typically separating initiatives into categories such as:
near-term / operational value
mid-term / growth and optimization value
long-term / transformational or strategic innovation value
This makes it ideal for AI portfolios, because AI initiatives rarely create value on the same timeline.
For example:
one AI initiative may reduce support costs this quarter
another may improve forecasting over the next year
another may be experimental but create major strategic advantage later
A horizon-based framework helps leadership avoid a common mistake in AI investment governance: judging every initiative only by short-term ROI.
From an agentic AI business solutions perspective, this is especially important because AI portfolios often include a mix of:
automation projects
copilots and agents
analytics and prediction models
innovation pilots
foundational data and governance investments
Some of these generate measurable savings quickly, while others create value through capability-building, competitive advantage, or future scalability. A horizon-based framework gives a balanced and executive- friendly way to assess all of them.
Why the other options are incorrect
A). a simple cost and benefit analysis
This is too narrow for a portfolio of AI initiatives with different maturity levels. It may help with individual projects, but it does not effectively balance short-term wins with longer-term innovation value.
C). the internal rate of return (IRR) function
IRR is a financial evaluation tool, but it is not the best structured portfolio framework for AI initiatives, especially where strategic and non-immediate benefits matter. AI value often includes intangible and capability-based outcomes that IRR alone does not capture well.
D). a prioritization grid
A prioritization grid helps rank initiatives, usually by factors like impact and effort, but it is not primarily a framework for evaluating ROAI over different time horizons. It supports selection, not full portfolio return evaluation.
Expert reasoning
When a question includes these ideas together:
portfolio of initiatives
different stages of development
immediate and long-term value
strategic innovation
the strongest answer is a horizon-based framework.
That is the best way to assess AI investments across short-term, medium-term, and transformational horizons without undervaluing strategic initiatives.

Explanation:
To improve performance # Move to a multi-agent architecture
To improve accuracy # Add a grounding data source
The current design uses a single agent and a single prompt to complete a series of tasks. That is often a bottleneck. When one agent is responsible for everything, it has to manage multiple steps, multiple reasoning modes, and multiple task transitions in one flow. This commonly leads to:
slower response times
task overload
incomplete outputs
reduced efficiency as complexity grows
Moving to a multi-agent architecture helps performance because tasks can be separated by function. For example:
one agent can handle task planning
another can retrieve domain knowledge
another can perform structured reasoning
another can prepare the final response
From an agentic AI systems perspective, decomposition improves throughput and execution quality. Instead of one overloaded agent trying to do everything, specialized agents handle narrower responsibilities. That often reduces latency in practical enterprise designs and improves the reliability of task completion.
This also addresses the symptom of incomplete results, because a multi-agent architecture can break a large workflow into smaller, controlled substeps.
Why "Add a grounding data source" improves accuracy
The agent struggles with domain-specific reasoning. That strongly suggests it lacks sufficient domain context during inference.
The best way to improve accuracy in this case is to add a grounding data source.
Grounding means giving the model access to trusted, relevant business knowledge at runtime, such as:
internal documentation
product specifications
policy manuals
knowledge bases
industry-specific reference data
This improves domain-specific reasoning because the model no longer relies only on general pretrained knowledge. Instead, it can anchor its responses in authoritative content.
From an AI business solutions standpoint, grounding is one of the most important mechanisms for improving:
factual relevance
domain accuracy
consistency
trustworthiness
explainability in enterprise contexts
When a model is inaccurate because it lacks business context, grounding is usually a better first fix than simply scaling model size.
Why the other actions are not the best fit
Add a prebuilt connector
A prebuilt connector helps with integration to systems and services, but it does not directly solve slow reasoning, incomplete output, or weak domain-specific reasoning unless the issue is specifically missing access to an external system. That is not the main problem described here.
Upgrade to a larger generative AI model
A larger model may sometimes improve reasoning quality, but it usually comes with higher cost and often slower response times, which works against the stated performance issue. It is not the best recommendation when the current agent is already slow.
Also, when domain-specific reasoning is the problem, grounding is usually more efficient and more controllable than simply choosing a larger model.
Expert reasoning shortcut
Use this exam logic:
Slow and overloaded single agent handling many tasks # move to multi-agent architecture Weak domain-specific reasoning # add grounding data source Need system integration # prebuilt connector Need raw generative capability increase, but can accept more cost/latency # larger model

If a Copilot Studio agent that uses Dataverse is returning poor or irrelevant results, the most likely issue is weak or inconsistent grounding data.
The best corrective action is C. Verify and cleanse the Dataverse data.
Why C is correct:
Grounded AI quality depends heavily on source data quality
Incomplete, inconsistent, duplicated, outdated, or badly structured Dataverse records will lead to poor answers Cleansing and validating the data improves both relevance and reliability without requiring unnecessary model changes Why the other options are not correct:
A). Use an adaptive card in Agent1Adaptive cards improve UI interaction, not grounding quality.
B). Add example user inputs to the training data of Agent1This may help intent handling, but the issue here is inaccurate grounding from Dataverse.
D). Retrain Agent1Copilot Studio grounding problems are usually solved by improving source data, not retraining the agent.

Adding fields to the opportunity summary does meet the requirement. Dynamics 365 Sales allows administrators to configure the fields Copilot uses when generating summaries for opportunities and other sales records. Microsoft explicitly directs administrators to configure fields for generating summaries and recent changes as part of Copilot configuration.
Microsoft also explains that although Copilot starts with predefined fields, organizations can add fields that are more relevant to their sales processes. Those selected fields become part of the information supplied to Copilot when generating the opportunity summary.
Therefore, modifying the fields directly changes the information included in the generated opportunity summary and qualifies as customization.