Business Readiness, Leadership, and AI Transformation

141. An AI readiness assessment must evaluate whether the organization can sustain AI initiatives. Which set of dimensions should it cover?

1. A Only the availability of GPU capacity
2. B Only the number of data scientists employed
3. C Only the current cloud spend
4. D People and skills, processes, technology and data foundations, and governance

142. Adoption of a deployed AI tool is low because staff do not trust its recommendations. Which intervention most directly addresses the cause?

1. A Reduce the cost of the underlying service
2. B Mandate use of the tool through a policy directive
3. C Increase the model's inference speed
4. D Improve explainability and provide training on how the tool reaches conclusions and where it should not be relied on

143. A company must forecast the ongoing cost of a generative AI application before committing to enterprise rollout. Which approach is most reliable?

1. A Model expected token volume per user interaction against the service's consumption pricing, using the AWS Pricing Calculator and pilot telemetry
2. B Estimate from the number of employees alone
3. C Assume cost scales linearly with the number of AWS accounts
4. D Use the training cost of the foundation model as the estimate

144. An AI readiness assessment finds strong technology and weak data governance. What should be sequenced first?

1. A Launching several pilots simultaneously to build momentum
2. B Procuring more compute capacity
3. C Hiring additional data scientists
4. D Data governance work, because model quality and regulatory defensibility both depend on it

145. Which structure most effectively spreads AI capability across business units without creating a bottleneck?

1. A An external consultancy owning all AI delivery
2. B A central team that builds every AI solution itself
3. C Fully independent teams with no shared standards
4. D A centre of excellence providing platform, standards, and enablement, with delivery embedded in the business units

146. Adoption of a deployed AI tool is low despite good measured accuracy. Which investigation should come first?

1. A Understand user trust and workflow fit, including whether the tool's output is explainable and fits how people actually work
2. B Increase the model's accuracy further
3. C Reduce the tool's operating cost
4. D Mandate use through a policy directive

147. Which communication approach best supports an AI transformation among staff who fear job displacement?

1. A State that no roles will ever change
2. B Be specific about which tasks change, how roles evolve, and what retraining is provided, rather than issuing general reassurance
3. C Emphasise only the cost savings the organization will achieve
4. D Avoid discussing the topic until deployment is complete

148. Which metric best indicates that an AI initiative has moved from pilot to genuine operational adoption?

1. A Sustained use in the daily workflow by the target population, with the business outcome metric improving
2. B The number of employees who attended the training session
3. C The number of model invocations in the first week
4. D The number of use cases identified

149. A company must build internal AI skills across technical and non-technical staff. Which approach is most effective?

1. A Role-differentiated learning paths, so leaders, domain experts, and engineers each receive relevant depth
2. B Hiring externally and not training existing staff
3. C A single organization-wide technical course
4. D Relying on staff to learn independently

150. Executive sponsorship for an AI programme is nominal, and decisions stall between departments. What is the most effective intervention?

1. A Reduce the programme's scope until no cross-department decisions are needed
2. B Secure an accountable executive owner with authority over the affected processes and budget, not only endorsement
3. C Increase the frequency of steering committee meetings
4. D Move the programme under the technology function

151. Which sequence best describes a sustainable path from experimentation to enterprise AI capability?

1. A Scale to every department first, then build governance
2. B Build a complete platform before attempting any use case
3. C Prove value in a bounded pilot, establish platform and governance, then scale to adjacent use cases while building skills
4. D Run indefinite pilots without committing to production

152. Which operating model suits an organization with strong central data capability but highly varied business domains?

1. A Outsourcing delivery entirely
2. B Fully autonomous domain teams with no shared platform
3. C A fully centralized team delivering all use cases
4. D A federated model with a central platform and standards, and domain teams delivering their own use cases

153. An organization must decide between hiring AI specialists, training existing staff, or partnering. Which factor should drive the mix?

1. A The relative salary cost of each option
2. B The preference of the technology function
3. C Whether the capability is durable and differentiating, in which case build internally, or transitional, in which case partner
4. D The speed of hiring in the current market

154. Which indicator suggests an organization is ready to scale beyond pilots?

1. A A signed contract with a major AI vendor
2. B A large number of pilots underway
3. C Enthusiasm from senior leadership
4. D Repeatable delivery patterns, a shared platform, governance in place, and at least one pilot with measured business benefit

155. Staff in an affected function are resisting an AI deployment. Which response is most likely to succeed?

1. A Involve them in designing how the tool fits the work, and address the concerns they raise about accuracy and accountability
2. B Escalate to their manager to enforce compliance
3. C Replace the resistant staff
4. D Deploy anyway and address concerns after adoption

156. Which measure of AI maturity is most meaningful to a board?

1. A The number of models in production
2. B The number of employees trained
3. C The proportion of core business processes measurably improved by AI, with realised benefit
4. D Total AI investment to date

157. An organization wants to build data literacy across non-technical staff. What content is most useful?

1. A The mathematics of neural networks
2. B How to interpret model outputs and uncertainty, what the systems can and cannot do, and when to escalate to a human
3. C Programming in Python
4. D The AWS service catalogue

158. Which practice most helps sustain momentum after an initial AI success?

1. A Publishing the measured result and the lessons, and having a visible pipeline of the next prioritized use cases
2. B Announcing the success widely without detail
3. C Pausing to write a comprehensive strategy document
4. D Moving immediately to the largest and most complex use case

159. A transformation programme must decide whether to automate an existing process or redesign it. Which consideration is decisive?

1. A Which option requires fewer approvals
2. B Which option is faster to implement
3. C Which option uses more AI services
4. D Whether the current process reflects constraints that AI removes, in which case redesign captures more value than automation

160. Which incentive structure best supports responsible AI adoption?

1. A Rewarding the largest AI budget secured
2. B Rewarding the fastest time to deployment
3. C Rewarding the number of AI projects launched
4. D Rewarding measured business outcomes and adherence to governance, rather than volume of AI deployed

161. A systems integrator is proposed to accelerate delivery. What should the contract prioritize to protect long-term capability?

1. A The lowest day rate
2. B The shortest delivery timeline
3. C The largest team size
4. D Knowledge transfer, documentation, and internal ownership of the resulting platform and data assets

162. Which communication cadence best supports an AI transformation across a large organization?

1. A A single launch announcement
2. B Communication only when a milestone is achieved
3. C Quarterly updates limited to senior leadership
4. D Regular updates showing measured progress, what was learned including failures, and what changes next for affected staff

163. An organization has strong pilots but repeatedly fails to move them into production. What is the most common underlying cause?

1. A Insufficient model accuracy in the pilots
2. B A shortage of AI vendors to choose from
3. C Lack of executive interest in AI
4. D Missing production foundations such as data pipelines, monitoring, governance, and clear ownership of the running system

164. Customer data is held separately by three business units with no shared access. What effect does this have on an AI initiative?

1. A Data silos improve model accuracy by keeping each unit's data specialised
2. B Data silos have no effect provided each unit's data is of high quality
3. C Data silos reduce the cost of an AI initiative
4. D Data silos limit the completeness of the training data and weaken the resulting model

165. An organization is assessing whether its data is ready to support an AI initiative. Which combination of factors should be evaluated? (Select TWO.)

1. A The number of database engines the organization operates
2. B Whether the data is accessible to the teams that need it
3. C The quality and completeness of the data
4. D The physical location of the organization's offices
5. E The programming language used by the data team

166. Several teams each claim a different definition of the same customer metric. Which foundation is missing?

1. A A data strategy with clear data ownership and agreed definitions
2. B A faster network between the teams' systems
3. C A larger data warehouse to hold every team's version
4. D An AI model trained to reconcile the definitions

167. An executive asks what infrastructure the organization needs before beginning an AI initiative. Which response is most appropriate?

1. A Assess what the initiative requires for data storage, access, and compute, rather than committing to a fixed platform first
2. B Build a data centre before evaluating any cloud option
3. C Defer the initiative until the organization has unlimited compute capacity
4. D Purchase the largest available compute capacity so the initiative is never constrained

168. A data sharing framework is being established for an AI programme. What is its primary purpose?

1. A To increase the total volume of data the organization stores
2. B To define who may access which data, for what purpose, and under what conditions
3. C To standardize the file formats used across the organization
4. D To reduce the cost of the organization's storage

169. Staff express concern that an AI initiative will eliminate their roles. Which leadership response is most appropriate?

1. A Restrict discussion of the initiative to the leadership team
2. B Communicate transparently about timelines, expected outcomes, and how roles will change
3. C Delay announcing the initiative until it is already deployed
4. D Reassure staff that nothing about their work will change

170. An AI initiative has technical backing but repeatedly stalls when it requires decisions from other functions. Which intervention is most appropriate?

1. A Move the initiative to a different technical platform
2. B Increase the size of the technical team working on the initiative
3. C Establish executive sponsorship and a cross-functional team with clear accountability
4. D Extend the initiative's timeline to allow for the delays

171. Which cultural barrier most commonly slows AI adoption in an established organization?

1. A The absence of a formal AI certification programme
2. B Risk aversion and fear of failure among teams asked to change how they work
3. C The cost of cloud compute capacity
4. D A shortage of available foundation models

172. An organization wants to build AI literacy quickly across a large workforce. Which combination of approaches is appropriate? (Select TWO.)

1. A Run targeted training programmes matched to each role's needs
2. B Require every employee to complete a technical machine learning course
3. C Run hackathons and proof of concept programmes so teams learn by building
4. D Restrict AI tool access to the technical teams
5. E Hire external consultants to perform all AI work

173. A process currently performed manually will be largely automated by an AI system. Which transition for the affected staff is most appropriate?

1. A Reassign them to unrelated work with no connection to the new system
2. B Reduce their hours in proportion to the work automated
3. C Move them into oversight and exception handling, where human judgement and empathy add most value
4. D Keep the manual process running alongside the AI system indefinitely

174. Which role do AI champions play in an enterprise transformation?

1. A They approve the technical architecture of each AI system
2. B They write the organization's AI governance policy
3. C They negotiate contracts with AI vendors
4. D They sustain momentum locally by advocating for adoption and supporting colleagues through change

175. An AI maturity assessment must cover more than technology. Which dimensions should it include?

1. A Technology capability only
2. B The number of data scientists employed
3. C The cloud services the organization has adopted
4. D People and skills, processes and governance, data readiness, and technology together

176. An organization's AI readiness assessment identifies a capability gap. Which response is appropriate?

1. A Hire externally for every gap identified
2. B Record the gap and reassess next year
3. C Define a progression pathway with specific interventions and a way to measure movement along it
4. D Deprioritize AI until the gap closes on its own

177. Which indicator suggests an organization is at an early stage of AI maturity?

1. A Reusable AI components are shared between teams
2. B A cross-functional governance body reviews AI initiatives
3. C AI capabilities are embedded in several business processes
4. D AI work happens in isolated pilots without shared standards, platforms, or governance

178. An organization must establish who is accountable for the quality of a dataset used by AI systems. Which practice applies?

1. A Recording the dataset's size and location
2. B Encrypting the dataset at rest
3. C Assigning data ownership with defined responsibility for quality, access, and lifecycle
4. D Replicating the dataset to a second location

179. An AI initiative is blocked because the data it needs is held by a business unit unwilling to share it. Which intervention is appropriate?

1. A Abandon the initiative
2. B Escalate each request individually to leadership
3. C Copy the data without the business unit's agreement
4. D Establish a data sharing framework defining permitted purposes, access conditions, and accountability so sharing is governed rather than negotiated case by case

180. An organization must decide what infrastructure foundations an AI programme requires. Which approach is appropriate?

1. A Defer infrastructure decisions until after the first deployment
2. B Derive the requirements from the intended use cases' data volumes, latency needs, and governance obligations
3. C Match the infrastructure another organization has published
4. D Provision the largest available capacity so nothing is constrained

181. A communication strategy for an AI transformation must address employee concerns credibly. Which approach is appropriate?

1. A Communicate what will change, when, and how roles are affected, including what is not yet decided
2. B Communicate only to managers and let them relay the message
3. C Communicate once at the start and then proceed
4. D Communicate only the benefits of the transformation

182. An organization must build AI capability among staff whose roles will change rather than disappear. Which approach is appropriate?

1. A Provide generic AI awareness training to everyone equally
2. B Replace them with staff who already have AI skills
3. C Reskill them for the parts of the role that remain and for the new oversight and judgement work the AI creates
4. D Wait until the change is complete before deciding

183. An AI centre of excellence is being established. Which purpose should it serve?

1. A Owning and delivering every AI initiative centrally
2. B Managing the organization's cloud spend
3. C Setting standards, sharing reusable components and practices, and supporting business units rather than owning every initiative
4. D Approving each business unit's technology choices

184. Which measure indicates that an AI transformation is taking hold within the organization?

1. A The size of the central AI team
2. B The number of AI tools the organization has licensed
3. C Business units independently identifying and delivering AI use cases using shared standards
4. D The number of employees who have attended AI training

185. An AI pilot has succeeded and must be scaled to the enterprise. Which consideration is most often underestimated?

1. A The cost of the foundation model at higher volume
2. B The training required for the central team
3. C The operational, governance, and integration work required to run the capability reliably at scale
4. D The time to provision additional infrastructure

186. A pilot must be designed so its results indicate whether the capability will work at scale. Which design characteristic is appropriate?

1. A Using the most capable model available regardless of production constraints
2. B Using representative data and real users rather than curated examples and a friendly audience
3. C Demonstrating the capability to executives rather than measuring it
4. D Using the smallest possible dataset to finish quickly

187. An organization has many AI pilots but few in production. Which cause should be investigated first?

1. A Whether the business units lack interest
2. B Whether the organization has the operational foundations, governance, and ownership needed to run a capability in production
3. C Whether the pilots used the wrong foundation models
4. D Whether the pilots were too small

188. An AI capability deployed enterprise-wide must remain available when a component fails. Which business consideration applies?

1. A Accepting any outage because AI is supplementary
2. B Defining what degraded operation is acceptable and what the business does when the capability is unavailable
3. C Ensuring the capability never fails
4. D Duplicating the capability with a second vendor

189. A feedback mechanism must be established as an AI capability scales. Which purpose does it serve?

1. A Recording which teams requested the capability
2. B Tracking the capability's infrastructure cost
3. C Measuring how many users have adopted the capability
4. D Capturing where the capability performs poorly so it can be improved and where it should not be relied upon

190. Which data foundation must be in place before an organization can train models on its own data?

1. A Data stored in the same Region as the model
2. B Data that is accessible, of known quality, and cleared for the intended use
3. C Data formatted as structured tables
4. D Data stored in a single database engine

191. Which approach helps staff trust an AI system they are asked to work alongside?

1. A Emphasising the system's accuracy figures
2. B Deploying the system without announcement
3. C Being transparent about what the system does well, where it fails, and how their judgement remains part of the process
4. D Requiring staff to use the system

192. Which sign indicates that an AI transformation is failing to take hold culturally?

1. A Staff ask questions about how the system works
2. B Staff continue performing the previous process alongside the new system rather than relying on it
3. C Staff report errors in the system's output
4. D Staff request additional training

193. An organization must decide how to structure its AI capability across business units.

1. A Balance a central function that sets standards and provides shared capability with unit-level teams that own their use cases
2. B Outsource all AI work to a vendor
3. C Centralize all AI work in one team
4. D Devolve all AI work to business units

194. Which measure indicates that AI literacy is improving across an organization?

1. A Business units independently identifying viable use cases and recognising unsuitable ones
2. B The size of the central AI team
3. C The number of employees who attended training
4. D The number of AI tools licensed

195. How should an organization address a role that AI will substantially change rather than eliminate?

1. A Replace the people in the role with new hires
2. B Leave the role unchanged until the change is complete
3. C Define the new shape of the role, including the oversight and judgement work, and reskill the people in it
4. D Reduce the role's responsibilities proportionally

196. Which leadership action most supports an enterprise AI transformation?

1. A Reviewing the technical architecture
2. B Approving the AI budget
3. C Visible executive sponsorship with accountability for outcomes rather than for technology deployment
4. D Selecting the organization's AI vendors

197. An organization must decide when an AI pilot is ready to scale.

1. A When it has demonstrated the intended outcome with representative users and data, and the operational foundations to run it exist
2. B When the pilot budget is exhausted
3. C When the executive sponsor approves
4. D When the technical implementation is complete

198. Which consideration most often prevents a successful AI pilot from scaling?

1. A The pilot's model not being the most capable available
2. B The pilot team being too small
3. C Insufficient cloud compute capacity
4. D The absence of the governance, operations, and integration work that production requires

199. How should an organization sequence the scaling of AI across business units?

1. A Scale to the largest unit first regardless of readiness
2. B Scale in the order units request it
3. C Start where the value is clearest and the foundations are strongest, then apply what is learned to the next unit
4. D Scale to every unit simultaneously

200. Which mechanism helps an organization avoid rebuilding the same AI capability repeatedly?

1. A A central approval process for new initiatives
2. B A record of each unit's completed projects
3. C A shared catalogue of reusable components, patterns, and evaluated models that units can adopt
4. D A quarterly review of AI spend

201. An AI capability deployed enterprise-wide must continue to deliver value as conditions change.

1. A Transfer the system to the operations team without ownership
2. B Monitor its outcomes against the business objective and retain ownership that can act when the value declines
3. C Review the system at the next annual planning cycle
4. D Freeze the system once it is working