Produce a 100-day plan: quick wins, foundational work (governance/evals), first agent rollout, security controls, and stakeholder cadence. Include a weekly operating rhythm and deliverables.
Create a moat analysis: data network effects, workflow embedding, switching costs, and compliance advantage. Provide 10 defensibility tests and a 12-month moat-building plan.
Define north-star metrics for AI in 2026: time-to-decision, cost-per-outcome, error-rate under policy, trust score, and automation rate with human satisfaction. Provide instrumentation plan.
Create a sovereign AI readiness checklist: data residency, model hosting options, key ownership, audit access, and supply-chain risks. Provide a staged maturity model from 0→4.
Design a provenance layer for internal RAG: citation IDs, document signing, version pinning, and ‘evidence packs’ in responses. Include UX and evaluation methods.
Write incident response runbooks for AI failures: hallucination spike, data leakage, tool misuse, latency blow-ups, and agent runaway. Include severity levels, comms templates, and postmortem format.
Create a data contract strategy for AI pipelines: schema/versioning, quality SLAs, lineage, and rollback. Include a ‘self-healing’ playbook and governance checkpoints.
Define a framework to decide when AI is a feature, a workflow, or a product line. Provide examples, pricing implications, and a roadmap template for 2026.
Design a finance AI roadmap: close automation, variance analysis, forecasting, and controls. Include policy for human sign-off and metrics tied to close time and accuracy.
Create a strategy for agentic CRM: lead research, email drafting, call summaries, next-best-action, and governance for customer data. Include ‘do-not-do’ rules and auditability.
Design an agent-enabled support strategy: deflection without hallucinations, verified knowledge, escalation, QA sampling, and customer trust metrics. Include a phased rollout and KPI ladder.
Propose a talent strategy: ‘forward-deployed engineers’ model, upskilling paths, incentives, and org design for AI product delivery. Include hiring profiles and interview rubrics.
