Mastering CI Maturity: Understanding Level 1 and How to Move to Level 2
In a previous article, I introduced a capability maturity model for organizational adoption and use of AI and other collaborative technologies (aka “collaborative intelligence” or CI).
This article focuses on Maturity Level 1, named “Exploring.” It will:
provide more detail about the observable characteristics at this level,
present strategies and actions that will enable organizations to move to Maturity Level 2.
Subsequent articles will focus on Maturity Levels 2-6. When this series of articles concludes, I will provide an assessment tool that helps organizations determine their current maturity level.
Characteristics of Organizations “Exploring” Collaborative Intelligence
Organizational Awareness/Knowledge of CI Technologies
Awareness of the range and capabilities of CI technologies is limited, especially among the senior leadership team. While leaders may be aware of generative AI (e.g., ChatGPT, Gemini, DALLE-3, etc.) due to its growing popularity, they are mostly unaware of other technologies that could improve organizational performance, such as computational intelligence, digital twins, or collaborative robots. Similarly, specific individuals or groups within the organization may have broader awareness, but they are the exceptions. Thus, the organization’s ability to see opportunities for CI-driven improvement across the organization is limited.
CI-Related Skills/Experience and CI-Specific Skills Management Processes
The organization has limited depth and breadth of CI-related skills and experience. While some individuals may have self-taught knowledge or have taken a course or two on specific CI technologies (e.g., prompt engineering), they are isolated. (The one exception is within an organization’s Data Science teams, which should house skills/experience with machine learning and other CI-applicable techniques.) The skills and experiences required to successfully redesign CI-enabled workflows, tasks, and jobs are also required, though not to the same extent (that is, they are required within whatever groups would be responsible for designing and implementing these changes).
The lack of training programs and standardized approaches to skill development means that CI skills/expertise remain siloed and inconsistent. There is no formal process to manage them, making it difficult to identify, develop, and leverage CI capabilities across the organization.
This lack of formal skills management hinders the organization's ability to systematically explore and implement CI technologies, leading to missed opportunities for innovation and the achievement of other benefits.
Influence of CI on Business Strategy
Leaders may be working to understand and anticipate the threats and opportunities posed by CI, but there have yet to be any CI-related changes to corporate or business unit strategies.
CI Technologies Deployed
Most uses of CI within the company are grass-roots explorations by individuals or ad hoc groups, with little or no coordination and no standardization of approach. The uses tend to be limited in scope (e.g., using generative AI to produce code snippets, emails, etc., or using Microsoft’s Copilot® to complete specific tasks while using Microsoft Office® applications). Any value produced for an individual or group is often difficult to replicate across the organization because such applications are neither standardized nor repeatable/efficient at scale.
CI-Enabled Business Processes, Job Designs, and Team Structures
To the extent that CI technologies are used to support the execution of specific tasks, they have been “bolted on” to existing business processes. No new or redesigned CI-enabled business processes have been introduced, and neither jobs nor organizational structures have been redesigned.
CI-Enabled or Embedded Products/Services
Senior leaders, product managers, and product teams may be considering how best to incorporate CI (especially generative AI) into their product/service offerings or actively working to do so, but none or very few new or enhanced offerings have been released.
CI-Aware Management Processes/Metrics
Since CI-enabled business processes do not exist, and no job descriptions reflect the use of CI, the related management processes and metrics are not necessary.
CI Governance Processes and Policies
The company may have issued guidelines for using generative AI in the workplace. Formal policies adhering to some or all characteristics of Responsible AI (RAI) (e.g., legal, safe, trustworthy, ethical, and unbiased) may be under development but have yet to be released.
Organizational Culture
While senior leaders and employees (each from their own perspectives) may be considering how organizational culture will have to change to support successful CI adoption, no steps have been taken to define or create this culture.
Data Models, Architectures, Workflows, and Management Processes/Policies
Data is often isolated, and its use is limited to systems that produce or consume it directly (e.g., a data visualization tool sitting on top of a CRM system). Data models may be inconsistent across business systems (e.g., customer records have different fields in different ERP or other enterprise systems). Data architectures are equipped to handle structured data only or primarily (e.g., it is difficult to capture/store semi-structured, unstructured, binary, and streaming data). Workflows that turn raw data into information and insights are not standardized and may be manual or require manual intervention. Management processes/policies may be limited in scope, ignored, or require manual execution/enforcement, and they have not been updated to reflect the changes required to support CI.
Computing and Cybersecurity Infrastructure
The organization's computing and cybersecurity infrastructure is not designed to support CI efficiently and effectively. Its computing infrastructure is adequate for existing operations but lacks the scalability and flexibility required to support advanced CI applications and CI-driven products/services. Cybersecurity measures often focus on compliance rather than comprehensive protection against sophisticated threats that exploit CI. Security protocols may not adequately protect non-traditional but sensitive data used in CI applications.
Key Strategies/Actions to Reach Maturity Level 2
Organizations at Maturity Level 1 have much to do to achieve higher maturity levels. However, rather than taking a broad-based approach to improvement, they should focus initially on enhancing their technical infrastructure and building the knowledge, skills, and experience required to prosper in the coming Age of AI. Regardless of what business strategies, process redesigns, and CI-driven products/services emerge in an organization’s future, every organization will need access to the technical infrastructure and human capabilities that will drive CI success.
Therefore, companies who want to move to Level 2 (and beyond) should take the following actions:
Harness their data.
Build organizational awareness of and experience with technologies beyond generative AI.
Conduct focused experiments using one or more CI technologies and turn the most promising experiments into operational systems.
Harnessing Data Assets
“Data is the new oil” is an oft-repeated analogy that helps executives understand and appreciate the value of their data assets. Just like oil…
Organizations must “discover” all the data they have or could obtain.
The more data organizations have, the better off they are.
The data must be extracted, processed, and refined to maximize its value.
The data can be used in many ways.
Companies at Maturity Level 1 must identify all their data assets, design a suitable and extensible data architecture that enables efficient and effective use of its data, and then build the infrastructure required to extract, store, refine, and utilize it.
Until its data is harnessed, it will be difficult for an organization to utilize it effectively and efficiently to support the development and use of CI-enabled systems and processes.
Build Familiarity with Other Intelligent Technologies
Generative AI is easy to use and has many valuable applications. However, it is just one of many technology categories that support human-machine collaboration. Over time, combinations of these technologies (called compound systems) will drive the most significant performance gains and create the most sustainable advantages.
Thus, companies at Maturity Level 1 should begin tracking developments in these areas, learn more about these technologies as they show promise, gain experience applying them to business problems or opportunities, and develop, acquire, or partner to build the requisite skill capacity to support enterprise-wide development, deployment, operation, and maintenance/enhancement of systems and the processes that utilize them.
While many companies realize this already, the war for talent in these technology categories is just beginning. Companies that lose this war will find it challenging to prosper in the coming era.
Conduct Experiments to Identify Promising Systems and Process Improvements
The power of Collaborative Intelligence technologies is improving rapidly, and the number of CI technology categories is expanding quickly. Such an environment creates a lot of strategic uncertainty. Thus, companies at all maturity levels should adopt an experimental approach to technology development, product/service development, and process improvement as they pursue CI-driven benefits.
A previous article described this experimental approach, and a section of a previous flipbook described how to manage the move from experiment to operational system, so I will not repeat that guidance here.
Executing this action plan will move an organization from Maturity Level 1 to Maturity Level 2 while providing a foundation for moving to still higher maturity levels.
Collaborative Intelligence is a Transformativ, LLC publication. If you’d like to learn more about how to become an AI-powered enterprise, please contact us here.