GETTY

Every single digitally adept enterprise has the potential to enhance operations with the correct implementation of artificial intelligence technologies. I use the word correct because there is a significant difference between utilizing AI for its own sake versus properly embedding the technology in order to fulfill a specific set of firm objectives. In fact, a broad business implementation of AI has the potential to do more harm than good. Without a proper plan for adoption, firms can experience the detrimental consequences of resource and funding misallocation. The result? The reverse of optimization, wasted time and ultimately negative ROI–factors that a fast-paced digital company cannot risk. Therefore, it is imperative to establish a strategic roadmap for accelerated AI adoption that digitally mature businesses can follow, implement and subsequently benefit from. First, the article will introduce an overview of key AI maturity models with a focus on the AI Pioneers Maturity Model, after which a recommended 5-point AI adoption roadmap will be presented.

A brief overview of key AI maturity models

Maturity models aid enterprises in focusing on their AI initiatives. These models should provide a useful framework for identifying AI’s potential strategic business impact, assessing an organization’s current AI capabilities and prioritizing investments toward AI technologies, skills and processes that are needed to boost readiness and achieve the desired outcomes.

Maturity models enable practitioners to determine whether or not a given enterprise’s AI ambitions are realistic or wishful hype. In addition, maturity models must be useful enough to capture the core of what AI offers without making the false implication that there exists a static body of best implementation practices.

Below are some previously published models and methodologies that were observed as part of the AI Pioneers Maturity Model:

1. The “Pragmatic AI” of “building blocks” framework: proposed by Mike Gualtieri of Forrester Research, this framework champions the acceleration of enterprise technology evolution via step-by-step adaptation of “building block” AI technologies like deep learning and machine learning.
2. Gartner AI Maturity Model: proposes maturity indicators to measure progress of AI adoption and fine-tune enterprise goals, as well as utilization of the “building block” framework above.
3. Adrian Bowles Maturity Model: this framework outlined by Adrian Bowles of Aragon Research is meant to assess how far-reaching an enterprise is in AI based on its current technology and skillset.
4. Enterprise Organization Maturity Phase Approach Model: a step-by-step maturity roadmap that begins with automation, followed by data centricity and digitalization and ending with reshaping the work environment to fully align with AI ambitions.
5. The Four Waves of the Intelligent Business by Workday: the cloud-based HR provider Workday breaks down AI maturity into four simple stages: automation, information, discovery and transformation.
6. Microsoft’s AI Maturity Curve: similar to Workday, Microsoft also decomposes AI maturity in four steps, beginning with understanding how to apply AI, digitalization, experimentation and preparing the business model with AI readiness in mind.
7. IBM’s Ladder of AI: lastly, IBM’s framework focuses on the early steps that an enterprise must take in order to prepare itself for AI adoption, such as recognizing data capabilities and augmenting analysis via machine learning methods.

A.I. Pioneers AI Maturity Model (Mark Minevich/AI Pioneers 2019)

The AI Pioneers Maturity model provides a framework for each level of AI Maturity, and outlines the relationship of an enterprise with AI at each stage. Advancement through these stages is not linear nor expected at a given adoption speed. Enterprises that deploy AI should approach the roadmap with a willingness to change speed and strategy throughout progression, and keep the model in mind as a measurement and benchmarking tool aimed at helping them move toward subsequent steps of maturity.

Relevant examples of this framework in action include the NBA utilizing AI to display key moments from basketball games, or TikTok powering its user recommendation algorithm with machine-learning technologies. Both companies have implemented a phased approach for artificial intelligence adoption as suggested by the AI Pioneers model. Furthermore, having already championed a data-centric mindset, both the NBA and TikTok were already technologically poised and mature from the get-go to correctly proceed with their respective AI ambitions, and more importantly, accelerated adoption. The NBA and TikTok are perfect examples of what today’s highest maturity level enterprises resemble, a matter that will be discussed subsequently.

Looking at Today’s Highest Maturity Level Enterprises

GETTY

In 2019, Dataiku (a firm that develops AI for enterprise solutions) conducted a survey entitled “AI Maturity Survey: Where Are We in the Path to Enterprise AI?” The results of over 350 poll respondents enabled Dataiku to establish concrete insights into the maturity level of digital enterprises. Among numerous factors extrapolated from the survey data, Dataiku found that the best equipped AI adoption organizations have the following in common:

1. Enterprises are already using some form of machine learning technology to augment processes.
2. Enterprises possess dedicated data units.
3. Enterprises institute some form of employee education to prepare for adoption.

To gain a better understanding of today’s highest maturity level enterprises, it is wise to review the metrics involved in adoption preparation. Statistics from Forbes reveals that 37% of AI industry leaders have poured more than $5 million into artificial intelligence technologies to optimize processes. On the basis of AI implementations, The Enterprisers Project shows that company institution of artificial intelligence technologies has increased by three-fold over the last 12 months, and 2 out of 3 large organizations will adopt AI in some form over the coming 24 months.

Each AI-centric enterprise has followed some form of an implementation roadmap in order to fulfill its artificial intelligence maturity objectives. The following information provides a 5-point guideline that digitally inclined enterprises can institute to fast-track AI adoption.

1. But First, Data

The first leg of the AI adoption journey requires businesses to already possess strong data analytics and metric-gathering capabilities. Artificial intelligence technologies, such as machine learning algorithms, must be fed a substantial and consistent stream of data in order to be useful. If enterprises are not data-geared, the power of AI vanishes, as it lacks the fuel and capacity to provide any meaningful insights to guide the decisions of firms.

Currently, there are ways to calculate the maturity level of AI adoption for enterprises, which can help point organizations in the right direction for implementation. One such tool is provided by Microsoft, which uses a survey-based questionnaire to gauge the adoption capacity of enterprises. Worldlink also provides an 8-step machine learning and artificial intelligence maturity guide to enable organization self-assessment. More importantly, the utilization of said maturity assessments in tandem for already data-centric enterprises is a surefire way to align organizations towards accelerated AI adoption and continual progress in maturation.

2. How Can (A)I Help You?

Having identified strong data analytics capabilities, firms must then determine how AI can help by establishing a rigorously formulated business question. At this moment, the specific intent for AI is made clear. For instance, firms can optimize operations with AI through automation of human resources, or implement machine learning algorithms to analyze employee satisfaction and productivity. Whatever the objective may be, enterprises must come up with a proposal as to how artificial intelligence can be helpful in order to save time and accelerate the roadmap to proper AI integration.

3. Testing The Waters

The next segment of the roadmap deals with tests and evaluations of candidate AI applications. Prior to official deployment, firms must evaluate the intended AI mechanisms on a micro level in order to minimize errors, harness the fullest potential of AI and ensure a smooth transition come launch time. This means designing unique experiments and modifying key performance indicators (KPIs) to measure progress and initiative efficacy. An example of a test could be A/B testing, whereby machine learning algorithms and employees carry out identical processes in a bid to empirically determine the extent of AI’s serviceability. Other quantitative tests include testing for accuracy, precision and speed.

In conjunction with the first round of experimentation, multiple-week-long test periods can be implemented to further examine and criticize the efficacy of the deduced methodologies as if firms were already in the short term post-launch environment.

4. Litmus Test Round Two And Establishing Concrete Use Cases

Following the initial round of tests would be a secondary experimental phase defined by longer test periods and the establishment of concrete use cases. During the secondary phase, test-periods would last for longer durations of time in order to best emulate the long term post-launch environment. The product of this final evaluative process would come in the shape of definitive AI use cases that could positively augment enterprise operations and consequently be used in the next phase of the adoption roadmap: amending the business model.

5. Amending The Business Model To Include AI

At this point in the AI adoption journey, enterprises must augment their business models in order to accommodate artificial intelligence. This step in the roadmap is crucial for relaying the implications of AI to all stakeholders involved, from investors to customers and elaborating upon the impact of the installation.

Launch Time And The Future

Once the business model has been amended, firms will have reached the proper level of maturity to correctly adopt artificial intelligence, hence finishing the long journey home. Lastly, as AI experiences persistent innovation over the coming years post-launch, firms that have implemented a strategic adoption roadmap will be well equipped to include the new technological advancements.

Final Thoughts

A data-centric mindset is imperative to successful roadmap completion. Proper enterprise acceleration of maturity for AI adoption means that companies should focus on optimizing their data units and preparing educational initiatives prior to commencing the implementation process. Organizations should also be dynamic in their approach, and be willing to amend their business strategy to incorporate artificial intelligence at every step of the way. This is the essence of a digitally mature AI enterprise.

Form:forbes.com