As technologists, we are always on the lookout for the next shiny thing. DevOps, Agile, Containerization, and Cloud Adoption have all been trending in recent years, and now Generative AI is joining that list. However, we need to be aware that like previous trends, Generative AI will require clarity of outcomes and organizational change management to be successful.
New outcomes are the key to sustaining any digital transformation. Without a clear line of sight to new outcomes, transformation programs are at risk of being defunded. It is important to remember that these outcomes are not technological outcomes, but rather line of business outcomes. A few examples might include achieving a deeper share of wallet, improving customer retention, or shortening sales cycles.
To ensure success with Generative AI in our organizations, it is vital to have a clear understanding of our expected outcomes and be open to adapting our processes accordingly. This starts with a common objective aligned with relevant stakeholders, followed by guiding principles for ethics, privacy, and transparency. Many organizations have published guidelines and principles for AI, such as the EU guidelines on ethics and artificial intelligence, IEEE, or several leading technology companies. If you have not done so, take the time to work with your internal and external stakeholders to build your vision and guiding principles.
"AI-mature organizations invest in foundational capabilities that will remain relevant regardless of what happens tomorrow in the world of AI" (Gartner Survey on Generative AI, May 7, 2024)
The greatest challenges that I hear from clients regarding Generative AI start with issues that are not easily automated. Data governance, authoritative source management, and data curation are at the core of these challenges. It is particularly important that organizations manage the bias in their models. Associated with bias management is the need to have audit and legal actively involved in reviewing the data management processes. Organizations must be able to demonstrate the appropriate due diligence regarding model data ingestion and model drift management, and these can only be done by establishing the appropriate processes. Often, we find that the challenging work of people and processes must be tackled to make progress. However, organizations do not have to take on the entire ecosystem.
By starting with an aligned objective, an organization can limit the effort to just the issues associated with those expectations. As is often the case in organizational change management, start small and celebrate the successes. Organizations must take the time to evaluate their needs and make informed decisions rather than blindly following the latest trend. Ultimately, people and processes are at the center of the work that must be done to be successful. Therefore, organizations must include organizational change management as part of their Generative AI strategy.
People are the only component of the change model that cannot be configured. You can configure technology and processes, but people must be aligned through common expectations. As Peter Druker famously shared, “Culture eats strategy for breakfast,” therefore take the time to align the objective and build the capacity to achieve those goals.
If you would like to know more about how to structure your AI organizational change management strategy, please reach out and connect with us below.
Comments