What is your sick care AI maturity model?

An AI maturity model is a framework that helps organizations assess their current AI readiness and capabilities. It provides a structured approach to measure the maturity of an organization’s use of AI and helps identify areas for improvement. The model typically consists of several stages, each with its own set of criteria and associated questions designed to assess an organization’s level of AI maturity. The number of stages and their specific criteria can vary depending on the model used.

In his book, The Four Steps to the Epiphany, Steve Blank described what has become the gospel of lean startup methodologies: Customer validation, customer discovery, customer creation and company building

The path to sick care digital/AI transformation is a bit shorter, but certainly no less difficult and plagued by failure: Personal innovation readiness, organizational innovation readiness and digital/AI transformation i.e. people, process, culture, and technology.

The people part is the hardest. On one hand, there is increasing data from high-quality experiments showing that AI really does improve task performance on many high-value work tasks by 20-60%, in fields ranging from coding to ideation to consulting. On the other hand, most CEOs are discouraging AI use as "they don't fully understand it." The problem is that understanding comes from experimentation. These firms won't learn, because they won't experiment, creating a vicious cycle.

Here are barriers to the dissemination and implementation of AI in healthcare.

Here are some ways to overcome them.

Examples of AI maturity models are:

1. Microsoft model
2. MI10
3. Gartner has released an AI maturity model that segments companies into five levels of maturity regarding an organization’s use of AI
4. Element AI has defined five dimensions (Strategy, Data, Technology, People, and Governance) for each stage (i.e., level), from exploring to experimenting, formalizing, optimizing, to transforming.
5. This study reviewed state-of-the-art studies related to AI maturity models systematically. It allows a deeper understanding of the methodological issues relevant to maturity models, especially in terms of the objectives, methods employed to develop and validate the models, and the scope and characteristics of maturity model development. The analysis reveals that most works concentrate on developing maturity models with or without their empirical validation. It shows that the most significant proportion of models were designed for specific domains and purposes. Maturity model development typically uses a bottom-up design approach, and most of the models have a descriptive characteristic. Besides that, maturity grid and continuous representation with five levels are currently trending in maturity model development. Six out of 13 studies (46%) on AI maturity pertain to assess the technology aspect, even in specific domains. It confirms that organizations still require an improvement in their AI capability and in strengthening AI maturity.
6. MITRE maturity model
7. The Health Management Academy

Some models are industry agnostic, while some are more specific, like sick care, considering the unique industry characteristics, ecosystems, regulatory constraints, and legal and ethical issues. In addition, they lack empirical clinical validation and sexutple aim results reporting.Eventually, there will be industry standards that will enable stakeholders to evaluate how one organization compares to another, similar to the HIMSS digital maturity levels.

Sick care AI maturity models are immature. We need better generalizable standards that include not just structure and process but outcomes as well.

The model you select should depend on the value it creates. Som questions to ask are:

1. Are the standards or pillars applicable?
2. Are they valid and predictive of success?
3. Can you compare your results with other comparable organizations?
4. What are your KPIs/OKRs and they some models better at achieving some than others?
5. How does the model help you measure the value created?
6. How effective and efficient are the models getting you from one stage to the next?
7. What is the cost/benefit of using or not using a particular model?
8. How does the model inform your strategy about whether to make or buy?
9. What is the business model of the organization that created the model?
10. How easy is it to use, disseminate and implement with stakeholders?

Many companies are struggling to derive value from GenAI because of a fundamental flaw in their approach: They think of GenAI as a traditional form of automation rather than as an assistive agent that gets smarter — and makes humans smarter — over time. These authors suggest a framework, Design for Dialogue, for reimagining their processes to mirror the back-and-forth collaboration of human dynamics to create an effective and adaptable human–AI workflow. At the heart of the framework are three primary components: task analysis, interaction protocols, and feedback loops.

As they evolve, in addition to helping improve AI governance, they will reflect an organization's evidence-based ability to deploy responsible AI that will improve patient quality and safety, operations and finance and become a competitive marketing tool.

Arlen Meyers, MD, MBA is the President and CEO of the Society of Physician Entrepreneurs on Substack and Editor of Digital Health Entrepreneurship