Better understanding artificial intelligence and uncertainty

Artificial intelligence (AI) and machine learning (ML) have emerged as two of, if not the leading, topics of discussion across various industries. To answer this growing need, IBM and Red Hat have announced availability of innovative products such as IBM watsonx and Red Hat OpenShift AI, respectively. This article outlines a simple strategy to help manage the changes associated with deploying AI/ML products that are on a course to reshape how the world works.

This article is written from the perspective of a Red Hat Technical Account Manager (TAM). A TAM’s unique role as the primary point of contact for a customer on Red Hat technical topics, combined with our advocacy for best practices, allows us to deeply understand customer needs and align those needs with Red Hat’s internal strategies. However, this does not resolve the prevalent uncertainty with AI/ML.

Red Hat, IBM and the AI ecosystem

Both IBM watsonx and OpenShift AI support AI/ML workflows. IBM's watsonx enhances AI development and deployment, while OpenShift AI uses cloud technologies to deploy and manage AI/ML workloads.

AI/ML, like other disruptive innovations before it, is the latest technology to both show promise and generate anxiety. A generation ago, the internet brought data to our fingertips anywhere, any time, but triggered worry across many industries. Today, the underlying fear is that AI/ML will destroy jobs by automating thousands of manual tasks.

Revolutionary or Transformational change

AI/ML will possibly make many jobs obsolete, but if it follows the trends set by earlier disruptive technologies, it will also create many new jobs while delivering fundamental global changes. These changes broadly fall into two categories: Revolutionary/Transformational and Evolutionary/Adaptive.

Transformational refers to a foundational or radical change; a fundamental shift where new technology dramatically replaces an existing technology, system or process. For example, streaming platforms such as Netflix replaced physical media like Compact Discs (CDs)/Digital Versatile Discs (DVDs), effectively eliminating many companies in the physical video rental industry like Blockbuster. Another example is online booking services replacing travel agents for airlines, car rentals and hotels.

Where transformational change is a wholesale, rapid shift, evolutionary change is more iterative and gradual, but no less disruptive to how we do business.

Evolutionary or Adaptive change

Evolutionary or Adaptive change leverages trending technology to enhance existing processes or systems, increasing efficiency and productivity while still maintaining the overall structure and goals. For example, telco companies began offering Voice Over Internet Protocol (VoIP) at the start of the 21st century, delivering these services at a fraction of the cost of older technologies. Likewise, digital banking revolutionized how people manage their finances. Apps such as Venmo, Zelle and online/mobile banking systems replaced routine tasks like cash transfers and check deposits, reducing in-person visits to banks and financial institutions for primarily specialized services.

As a TAM supporting the telecommunications and enterprise sector, I often contemplate how telco companies will deploy Red Hat's AI products and solutions. Is the industry going to experience change via a transformational, adaptive or a mix of the two?

On one hand, Red Hat OpenShift AI and Red Hat Enterprise Linux (RHEL) AI are rapidly gaining momentum, offering advanced capabilities that could significantly enhance operational efficiency and service delivery. On the other hand, some telco companies prefer a more cautious approach to continue with their existing technology.

To support this more cautious approach, Red Hat is extending the Extended Life Cycle Support (ELS) phase on many traditional IT products, including RHEL and OpenStack. The more cautious telco companies appear to be maintaining their current deployments, founded on these ELS products, while closely watching how their competitors implement AI-enabled technology before making strategic decisions.

Just like earlier technology revolutions, successfully navigating the AI era requires us to manage these new  changes.

The following section introduces a technique to manage transformational and adaptive changes as AI deployment and adoption grow, offering practical steps to navigate the complexities and opportunities of the AI era.

Managing change through ADKAR

ADKAR is a popular change management model that can be used to guide individual and organizational change. It is a goal-oriented, structured approach that emphasizes the importance of facilitating individual transitions, recognizing that organizational success depends upon each person’s ability to adapt to change. Developed by Jeff Hiatt, the founder of Prosci, ADKAR stands for:

• A - Awareness
• D - Desire
• K - Knowledge
• A - Ability
• R - Reinforcement

Each component in the model represents a stage or a milestone that individuals must achieve for successful change implementation. The strength of the ADKAR lies in its sequential process, which enables a smooth transition from one phase to the next. For example, "Desire" cannot increase without first increasing "Awareness". Similarly, "Knowledge" cannot grow without "Desire," and "Ability" develops only after "Knowledge" is acquired, and so forth.

Here is a comprehensive breakdown of each component of the ADKAR model, with recommendations for how individuals and organizations can manage change as they adopt AI.

1. Awareness

Objective: The goal of increasing awareness is to understand the changes needed to succeed in the AI era by recognizing the shifts in technology and market dynamics and identifying how these changes may affect your role and the organization.

Actions:

• Understand the relevance of AI/ML by answering key questions:
  • What is the current Red Hat product deployment, and how do the AI/ML-enabled products fit in that workflow?
  • Will AI deployment enhance operational efficiency?
  • Do current AI products meet business goals, technical challenges and customer needs? If not, what are the gaps that need to be addressed?
• Stay informed by reviewing management communications, staying curious about Red Hat OpenShift AI and Red Hat Enterprise Linux (RHEL) AI , and exploring new features and applications
• Engage in webinars and conferences. Read white papers and articles to stay updated on AI/ML trends and developments in cloud computing. Reviewing authoritative journals like Institute of Electrical and Electronics Engineers, and industry websites like TechCrunch can be particularly helpful
• Research industry trends such as AI/ML’s impact on telco and other industries, and assess business implications by linking use cases to improvements in operations like automation and resource management

2. Desire

A strong awareness of AI will naturally lead to an increased motivation to learn and apply AI/ML technologies.

Increasing desire to support change helps align individual or team goals with business strategies leading to improved efficiency in processes and workflows, and can contribute to revenue growth and personal career advancement.

Actions:

• Define clear objectives for learning AI/ML, such as becoming a subject matter expert (SME) or leveraging AI/ML to optimize OpenStack/OpenShift deployments
• Identify motivators like career advancement, technical curiosity or client satisfaction
• Engage with peers and industry experts passionate about AI/ML to stay motivated
• Align learning with long-term career goals such as specializing in AI/ML or leading future AI/ML projects
• Network through AI/ML communities and user groups
• Seek mentorship from experts to guide your journey

3. Knowledge

Knowledge in the ADKAR process involves acquiring information and education to adapt, with a strong desire to change driving the process.

Actions:

• Study technical documentation on AI/ML fundamentals and advanced topics, specifically tailored to OpenStack/OpenShift and cloud computing
• Enroll in relevant training classes, webinars and educational events
• Take courses (such as RHOAI / AI267) to deepen your understanding of AI/ML
• Assess insights from executives, training and learning teams to identify knowledge gaps
• Join workshops and bootcamps focusing on developing practical AI/ML applications using Red Hat products

4. Ability

Ability refers to the practical application of the knowledge acquired to perform tasks associated with change and implementing new requirements.

Actions:

• Apply AI/ML techniques to solve specific problems or optimize processes within OpenShift/OpenStack
• Collaborate with cross-functional teams on AI/ML projects to gain hands-on experience
• Lead or participate in proof-of-concept (POC) projects for AI/ML in OpenStack/OpenShift

5. Reinforcement

Reinforcement involves embedding these changes into team or company culture  so it becomes a permanent part of operations.

Actions:

• Foster a continuous learning culture in AI/ML among peers
• Document and share successes and lessons learned
• Measure AI/ML impact with metrics
• Establish support systems, such as a shared knowledge repository and stay updated on training resources
• Reward and recognize progress and milestones

Wrap up

Use  the ADKAR model to systematically assess the relevance of AI/ML for your specific use case. This framework helps you strategically apply AI/ML methodologies and tools. Develop a comprehensive approach to mastering AI/ML technologies and driving impactful change in the team and ecosystem by progressing sequentially through the stages of Awareness, Desire, Knowledge, Ability and Reinforcement.

Learn more about the Red Hat AI

Blogs

• Elevating AI Journey with TAM Services: Engage with a Red Hat Technical Account Manager and learn more about Red Hat's AI Strategy
• Overview of Red Hat OpenShift AI
• Current AI activities

Training materials

• AI267: Consolidated AI learning platform
• AI262: Introduction to Red Hat OpenShift AI
• AI263: Red Hat OpenShift AI Administration
• AI264: Creating Machine Learning Modules with Red Hat OpenShift AI
• AI265: Deploying Machine Learning Models with Red Hat OpenShift AI
• AI266: Automating AI/ML Workflows with Red Hat OpenShift AI