Case: Scaling AI Without Losing Data Control
An insurer with three AI features in production and nine requests waiting must decide what domains own, what the platform enforces, and which features to admit first.
-
-
Case: AI Investment Under a Budget Reduction
A fictional insurer must cut a tenth of its technology budget, which falls on 37.5% of what can move. Five engineers, an overspent error budget, and an AI pilot at 45% adoption shape the choice.
-
Leading an AI Incident: Containment to Resumption
A chat assistant leaks another household’s details, promises coverage the policy does not give, and issues credits above its limit. Weigh containment options by their business cost, then notify, communicate, and resume with written criteria.
-
Leading Cross-Functional AI Transformation
Six AI efforts ran across a fictional distributor, each owned by a different function. The VP of AI Transformation decides who owns each use case and process, what happens to released capacity, and which decisions stay with the enterprise.
-
Platform Engineering Leadership for AI Adoption
A platform leader has 84 engineer-months and seven AI requests. Decide scope, standards, developer tooling, shared services, and reliability and cost investment.
-
Data Platform Leadership Decisions for AI
A new data platform leader faces an AI assistant request, ownership demands from business teams, and a cost cut. Work through the funding, ownership, readiness, and cost arithmetic behind each choice.
-
Deciding When an AI Pilot Is Ready to Scale
A fictional retailer’s pilot shows a 40.5 percent time saving. Corrected for case mix, audited for harmful errors, and priced per case, it supports a gated redesign, not company-wide use.
-
Measuring AI Productivity: Activity vs. Value
A usage dashboard says 78 percent adoption and 45 percent more changes. After baselines, comparison groups, and rework, the gain is a quarter of that. Here is how to measure AI productivity and decide on renewal.
-
Leading AI-Assisted Engineering Adoption
AI coding tools make writing code cheaper but leave verification costs in place. A worked review-capacity example, rollout stages, skill risks, and delivery measures show how a leader can decide when to expand.
-
AI Decision Authority and Human Approval
A fictional freight forwarder first requires a person to approve every AI action, then watches the queue break on a storm day. Match authority to consequence and reversibility, size approval capacity, and name who answers for each decision.
-
Redesigning Work Before Automating It with AI
A fictional distributor’s AI proposal would have sped up the part of a credit process that was already fast. Analyzing tasks, decisions, and handoffs moved the investment elsewhere.
-
Cross-Functional Transformation Review
A fictional insurer’s AI program is on budget, under a renewed security exception, and used on 31 percent of intakes. How a cross-functional review turns three accurate views into decisions.
-
AI Ownership Boundaries Across Data and Platform
A fictional retailer’s refund model held roughly three times as many requests after a data field changed meaning. Ownership maps, contracts, incident authority, and escalation decide how fast such harm stops.
-
Capacity Allocation for Platform, Data, and AI
Three directors each submit a reasonable plan, and all three depend on the same two engineers. How to count capacity by skill, order dependencies, staff, and reallocate.
-
Building a Leadership Team and Succession Bench
A VP of Data Platform loses her best director and finds no ready successor. How to hire and develop directors, make peers accountable to each other, and judge readiness from evidence.
-
Leading Through Directors
A fictional VP reversed a quarter of her directors’ decisions. Decision rights, goals tied to business results, monthly reviews, and escalation keep delegated work delegated.
-
AI Supplier Contracts and Exit Conditions
A pilot passed its tests, and the supplier’s standard agreement arrives with a three-year commitment. Worked numbers show which clauses decide model changes, data use, service levels, and the real cost of leaving.
-
Building an AI Investment Case That Can Be Tested
A pilot cannot rescue a case whose baseline, costs, or thresholds were never written down. Measure review time first, count every cost, and set the stop rule before the results arrive.
-
Choosing an AI Transformation Portfolio
A fictional insurer ranked forty-one AI ideas by average score. Tracing the workflow, shared dependencies, and non-AI alternatives changed which ones to fund, delay, and stop.
-
Budgeting People, Vendors, and AI Consumption
A platform group faces a mid-year cut after hiring slips and AI usage doubles. Separate fixed, stepped, and usage costs, forecast ranges, and protect commitments.
-
Linking Platform Investment to Business Outcomes
A fictional grocer spent 9.6 million on a data platform while revenue grew 14%. Estimate what the platform added, separate early signs from results, and decide under uncertainty.