After two years of excitement around generative AI, companies are shifting their focus to creating tangible value from the technology. A McKinsey Global Survey shows that 65 percent of companies now use gen AI, double the rate of last year. While investment continues to rise, many companies are still in the early stages, with limited impact. Data executives have been drafting gen AI strategies to integrate AI into tech stacks, but the true value lies in creating robust operating models. These models guide AI use across organizations, from staffing and organizational structures to technology development and compliance, ensuring measurable business results. A common challenge in deploying gen AI is the "tech for tech" trap, where resources are spent without clear business objectives, or a "trial and error" approach, which risks inefficient implementation.
A successful gen AI operating model requires a balance of strategic planning and flexibility. Chief data officers play a key role in driving this, often forming dedicated gen AI teams or incorporating AI capabilities into existing IT structures. Organizations that use a component-based model to integrate gen AI into their enterprise architecture can quickly scale operations, as seen in European banks and investment firms. Proper data management and governance, especially with unstructured data, are essential for ensuring AI systems retrieve and process the right information. Additionally, risk and compliance governance must be central to any gen AI deployment, addressing potential issues like hallucinations and data leaks. By creating structured, adaptable models, companies can unlock gen AI’s potential while maintaining operational security and alignment with business goals.
