Task 4.1: Assess AI business readiness and maturity.

• ATechnical infrastructure, by adding more compute capacity for AI workloads

  Incorrect

  The assessment found infrastructure is already modern. More capacity does not resolve the disagreement that is blocking funding.

• BData quality, through a cleansing program

  Incorrect

  The data is already well governed. A cleansing program addresses a gap the assessment did not find.

• CLeadership alignment on the goals AI should serve

  Correct

  AWS guidance on readiness stresses aligning stakeholders on the goals set in the envision phase, surfacing their concerns and educating senior executives on AI's strategic potential. Without that alignment nothing gets funded.

• DWorkforce skills, by sending the analysts to advanced machine learning courses

  Incorrect

  The analysts are already enthusiastic; more training does not create the executive agreement needed to fund any initiative.

Readiness spans leadership, data, culture, technology and governance, and the weakest dimension sets the pace. Here every dimension looks ready except the executive one, so aligning leaders on what AI should achieve comes before further investment elsewhere.

• ALevel 1, Envision

  Incorrect

  Envision is the stage of building awareness and exploring use cases. The company has moved past it to hands-on pilots with dedicated teams.

• BLevel 2, Experiment

  Correct

  The Experiment level's criteria are active pilots and proofs of concept, dedicated cross-functional teams, a structured internal training program, selected models and tools, and initial governance and data frameworks.

• CLevel 3, Launch

  Incorrect

  Launch requires production-ready solutions delivering measurable business outcomes. None of the company's solutions is in production.

• DLevel 4, Scale

  Incorrect

  Scale means widespread use across multiple departments with reusable components and automated governance, which is far beyond the current state.

AWS's maturity model has four levels: Envision, Experiment, Launch and Scale. Matching the organization's current state to each level's criteria shows where it stands and which focus areas come next.

• AForm cross-functional squads with IT, business, security and AI experts

  Correct

  AWS recommends forming cross-functional generative AI squads with clear roles to lead experimentation; they later become the basis for a center of excellence.

• BBuild automated CI/CD pipelines and infrastructure as code for AI solutions

  Incorrect

  Production-grade pipelines are part of moving from Experiment to Launch. Building them before any pilot exists invests ahead of need.

• CSecure budget and people for a few small proofs of concept

  Correct

  Allocating resources for small-scale proofs of concept is one of the recommended steps to move from Envision to Experiment.

• DCreate an internal marketplace of reusable AI components for all teams

  Incorrect

  Inner-source practices and a component marketplace are recommended for moving from Launch to Scale, when there are proven assets to reuse.

• EFormalize the enterprise-wide AI operating model and RACI matrix

  Incorrect

  Formalizing the operating model across the organization is a step toward Scale, after production systems exist.

Each maturity level has its own transformation steps. From Envision, the next moves are practical: cross-functional squads, prioritized use cases, a budget for small proofs of concept, preliminary governance and success metrics. Platform and operating-model investments come later.

• AA technology gap, because the lab needs more powerful foundation models

  Incorrect

  The proofs of concept succeeded, so model capability is not what stops them. They stall at the hand-over between teams.

• BA people gap, because the lab needs more data scientists

  Incorrect

  The lab already builds successful proofs of concept. More builders would produce more stalled pilots.

• CA process gap, because no hand-over to production is defined

  Correct

  AWS's maturity guidance lists documented hand-over processes for successful proofs of concept that move to production as an operations activity of the Experiment level.

• DA governance gap, because the bank needs a new responsible AI policy

  Incorrect

  A responsible AI policy is important, but the stated problem is an undefined transfer of ownership, documentation and testing between teams.

Capability gaps fall into people, process, technology and governance. Diagnosing which one blocks progress matters: here the work itself succeeds, but there is no agreed process to move it into production, so more people or better models would not help.

• ASeveral proofs of concept are running with small, dedicated teams

  Incorrect

  Active pilots and proofs of concept with dedicated teams describe the Experiment level the board wants to know it has left.

• BProduction AI solutions are delivering measurable business outcomes

  Correct

  AWS lists production-ready generative AI solutions that deliver measurable business outcomes among the criteria for the Launch level.

• COperational controls include automated monitoring and alerting

  Correct

  Launch-level criteria also include established operational controls with automated monitoring and alerting systems.

• DTeams are attending AI awareness workshops and industry conferences

  Incorrect

  Awareness training, workshops and conferences are Envision-level activities.

• EA self-service library of reusable AI components serves all departments

  Incorrect

  An organization-wide self-service library of reusable components is a Scale-level criterion, beyond Launch.

Each maturity level has observable criteria. Launch is reached when solutions run in production with measurable results and with the operational controls that production requires.

• AEnvision, the first level

  Incorrect

  Envision is about building awareness and exploring use cases, far behind the group's current state.

• BExperiment, the second level

  Incorrect

  Experiment is characterized by pilots and proofs of concept, not dozens of production applications.

• CScale, the fourth level

  Correct

  Scale-level criteria include widespread adoption across departments, an enterprise-wide infrastructure and tooling ecosystem, a reusable component library with self-service access and automated governance.

• DLaunch, the third level

  Incorrect

  Launch covers a select few production applications; widespread, self-service adoption goes beyond it.

Maturity models help organizations know where they stand. Matching observed characteristics against each level's criteria shows that this group has reached the Scale level.

• AIntegration between AI solutions and legacy core systems

  Correct

  AWS's guidance on AI-powered application development lists legacy systems integration in the technology and tools component, which identifies the integration points and data requirements AI needs.

• BEvaluation metrics for the utility's AI technology partners

  Incorrect

  Partner evaluation metrics matter when partners underperform; here the blocker is the connection to internal systems.

• CEthics policies that govern how AI uses operational data

  Incorrect

  Ethics policies are part of AI governance, but nothing in the scenario points to a policy gap stopping the pilots.

• DBusiness impact reporting for the pilots already completed

  Incorrect

  Reporting would show the pilots' value, but it would not let them reach the systems they depend on.

Readiness assessments should look beyond models and data samples. AI that cannot connect to the systems where work happens cannot reach production, so integration with legacy systems is a capability to plan for early.

• AEngage AI experts and consultants to explore use cases

  Correct

  AWS lists engaging with AI subject matter experts and consultants, exploring potential use cases and business benefits, and building knowledge among the key activities of the Envision level.

• BCreate reusable AI capabilities and components for all teams

  Incorrect

  Creating reusable capabilities and components belongs to the Scale level.

• CIntegrate an ITIL framework for AI operations and support

  Incorrect

  ITIL integration supports production operations at the Launch level.

• DFormalize a RACI matrix for scaled AI operations

  Incorrect

  Formalizing the operating model and RACI is a Scale-level activity.

Readiness starts with knowledge. Bringing in expertise to explore use cases and benefits is an appropriate first step for organizations at the beginning of their AI journey.

• AThe Scale phase, which expands production pilots

  Incorrect

  Scale expands proven pilots; it does not come before launching them.

• BThe Align phase, which identifies gaps, dependencies and concerns

  Correct

  The Align phase focuses on identifying capability gaps across the six perspectives, cross-organizational dependencies and stakeholder concerns, to improve readiness and alignment.

• CThe Launch phase, which delivers pilots in production

  Incorrect

  Launch delivers pilots; the gap analysis should come first.

• DThe Envision phase, which identifies transformation opportunities

  Incorrect

  Envision has already identified opportunities; the next step is alignment and readiness.

Skipping alignment is a common cause of stalled transformations. A deliberate look at capability gaps, dependencies and concerns prepares the organization before pilots begin.