How should a team start a gen AI initiative?
Why A is the answer
With a business need and constraints
Business requirements and technical constraints should shape the initiative, rather than a tool chosen first.
In detail
- Start with a business need and its constraints.
- Technology choices come after.
Use case
An illustrative scenario, not a real incident.
A team starts from slow claims handling, and its budget and data limits, rather than from a tool.
See it in different ways
flowchart LR A["Business need"] --> B["Choose a solution"] B --> C["Integrate"] C --> D["Measure impact"] D --> E["Scale or adjust"]
sequenceDiagram participant L as Leader participant T as Team participant B as Business L->>T: Need and constraints T->>L: Metrics, set before launch T->>B: Pilot B-->>L: Measured result
flowchart LR S0["Business need"] --> S1["Constraints"] --> S2["Solution choice"]
Key point: Start from the business need.
Why the other options are wrong
- B. With the newest model available
- A model is not a goal.
- C. With a company-wide rollout
- A big rollout comes after proof.
- D. With a custom chip design
- Chip design is unrelated.