A Pragmatic Approach to Implementing AI: Harnessing the Potential of Your Data Assets

We were delighted to have Joakim Wahlqvist , Group Head of Data & AI with Sogeti present at the recent Analytics Institute Dublin Analytics & AI Showcase. Joakim explored “A Pragmatic Approach to Implementing AI: Harnessing the Potential of Your Data Assets” looking at how organizations can strategically unlock the value of their data through AI.

Key Messages:

AI has been evolving for decades through a collective exploratory and innovative spirit that has resulted in a range of technology that is widely accessible to individuals and organisations. It’s not going away, and the promise of innovation is growing exponentially. But as with all innovation, we know from experience that elements of risk will arise. So it makes sense for organisations to create an exploratory and collaborative culture between business and IT to understand how AI technology will provide them with benefits, and also explore mitigation for the risks they encounter along the way.

?

We recommend a pragmatic approach, where organisations firstly seek to understand the business value that can be achieved through the application of AI to business challenges, and how that business value aligns to business goals. Every organisation is unique, so it follows that the starting point and order of developing the core capabilities for AI will also be unique and dependent on business context, existing capability level and business priority.

We view the following capability pillars as core for AI: ?

1. Business Capability (Ideation Process, AI Awareness and Training, AI Governance, AI Enablement, Business Adoption, AI Councils, Value KPIs,etc)
2. AI Readiness (Business Use Cases, Traditional AI, Trusted AI, Synthetic Data, Data Enrichment, Vector Embedding, Prompt Engineering)
3. Delivery (Change Management, ML Engineering, Enterprise Architecture, Cloud Engineering, Delivery Teams, Data Engineering, AI Engineering)
4. Data Readiness (Data Governance, Data Lineage, Data Engineering, Data Quality, Data Catalog, Visualisation, Data Products)
5. Foundational Capability (GenAI Framework and Software Development Kit, Tools, Guardrails and Scoring, Cloud Landing Zones, Model Hub, RAG, Agentic Framework)

?

These pillars are important but learning through implementation will be equally valuable! So a focused effort to start “small”, get going and develop products of value in iterations will set you on the right trajectory to develop scale. It might feel daunting to start this on your own, if AI implementation is new to your organisation, so you might consider an intelligent enterprise partnership that can bring years of experience in this field to guide you to develop these capabilities – that’s us, Sogeti!

?