Question 1

What is an AI strategy, and why does a business need one?

Accepted Answer

An AI strategy sets out where AI will create value for the business, what it will take to deliver, and how success will be measured. Without one, organisations accumulate disconnected pilots that never reach the P&L. A good strategy ties a small number of prioritised use cases to business outcomes, the data and ERP foundations they need, and a funded roadmap the board owns.

Question 2

How do you build an AI strategy that delivers ROI?

Accepted Answer

Start from the P&L, not the technology. Identify where decisions and processes drive cost, revenue and risk; score candidate use cases on value and feasibility; fund the few that can scale; and measure each against the business case it was approved on. Most returns come from embedding AI in existing workflows rather than standalone tools.

Question 3

Should we build or buy AI?

Accepted Answer

Decide per use case. Buy when the need is standard and speed matters. Build where AI differentiates you, data is highly sensitive or you need to own the IP. For most enterprises the answer is a hybrid: commercial platforms extended with your own models, data and integration, governed by a clear target architecture.

Question 4

What is AI governance, and who should own it?

Accepted Answer

AI governance is the set of policies, controls and accountabilities that keep AI safe, compliant and aligned to business goals, covering data, model risk, transparency, security and regulation. The board sets risk appetite; an executive sponsor owns delivery; a cross-functional group runs day-to-day controls. It works best designed in from the first use case.

Question 5

What role should the CEO and board play in AI?

Accepted Answer

AI is now a board-level topic because it changes cost structures, risk and competitive position. The CEO and board should set the ambition, approve the investment case, own the risk appetite and hold leaders accountable for value realised, while building their own AI fluency so they can challenge proposals with confidence.

Question 6

Why do most AI projects fail to scale beyond the pilot?

Accepted Answer

Usually for business reasons: pilots are not tied to a funded business case, data and ERP foundations are not ready, AI sits beside processes rather than inside them, ownership is fragmented and governance arrives too late. Fixing the sequence (value, foundations, operating model, governance) is what moves AI from pilot to scale.

Question 7

How do you build AI capability inside an organisation?

Accepted Answer

Run AI as a use-case portfolio with stage-gate funding, make your data AI-ready (data products, RAG and vector search), set a model strategy, stand up an AI platform with LLMOps, evaluation and guardrails, govern by design against the EU AI Act and ISO/IEC 42001, and industrialise adoption through an AI CoE and copilots embedded in workflows. Capability grows fastest when it is applied to the use cases that matter in your industry.

Question 8

How do you prioritise AI use cases and build an AI roadmap?

Accepted Answer

Assess opportunities across functions and rank each one by value, feasibility and risk. Fund the few that score highest and can scale, then set out a practical 90-day plan to prove value and a 12-month AI roadmap to scale it. The roadmap should name owners, data and ERP dependencies, and the measures each use case will be judged on.

Question 9

What does an enterprise AI architecture include?

Accepted Answer

A target AI architecture connects large language models and other AI services, an orchestration layer for agents and workflows, governed enterprise data, and the business applications and ERP where decisions are made. It also covers integration, identity and security. An AI readiness assessment of data, ERP, cloud, integration and security shows the gap between today and that target.

Question 10

What is an AI operating model, and do we need an AI centre of excellence?

Accepted Answer

An AI operating model defines how people, processes and technology change to use AI every day: who owns use cases, which roles and skills are needed, how ideas move from pilot to production, and how adoption is measured. A small AI centre of excellence (CoE) sets standards and shares expertise, while business teams own the outcomes. It is what moves AI from pilots to enterprise-scale deployment.

Question 11

How is AI changing M&A and private equity due diligence?

Accepted Answer

Investors now assess a target’s AI exposure and upside alongside its technology risk: the maturity of its data and ERP platforms, the disruption risk to its model and the credible AI value creation plan after close. Integration and separation planning is also shifting earlier, because AI benefits depend on bringing data and platforms together. This is a core focus of AumentoAI’s work with investors.

Jeet Poptani

Enterprise Transformation | AI Advisory & Leadership

Delivering ERP Platforms and AI Capabilities for Enterprises & PE Firms

Founder, AumentoAI
Digital Transformation Officer
B.Tech, MBA, London Business School

From the deal table to one integrated platform.

M&A

Private equity

Technology integration

Run AI as a use-case portfolio.

Set the model strategy.

Stand up the AI platform and LLMOps.

Govern and secure by design.

Industrialise adoption.

Carlsberg

Jaguar Land Rover

Diageo

AstraZeneca
Enterprise AI Platform Strategy
IMPLEMENTED