All Blog Articles

A Principles-Based Approach to AI Implementation

August 10, 2026 – Shingo Institute

Executive Summary

Artificial intelligence (AI) represents one of the most significant technological developments of our era, and its potential extends far beyond automation. AI can accelerate learning, improve quality, enhance decision-making, strengthen knowledge sharing, and increase an organization’s ability to create value. At the same time, AI introduces risks related to data quality, bias, privacy, governance, overreliance, and unintended consequences.

From a Shingo perspective, the critical question is not whether AI is good or bad. The more important question is, “How can leaders use AI in ways that are consistent with enduring principles?” Technology does not create culture. Leadership choices, management systems, and daily behaviors create culture. AI simply amplifies the strengths and weaknesses already present within an organization.

Based on the Shingo Institute’s third insight of organizational excellence—Principles Inform Ideal Behaviors[1]—this whitepaper explores how a principles-based approach to AI implementation can accelerate improvement efforts while preventing potential destructive consequences.

Strategy

The Shingo Institute observes many of the world’s highest-performing organizations. These organizations demonstrate that sustainable excellence is the result of systems driving behaviors informed by principles. While tools and technologies continue to evolve, the principles that govern organizational excellence remain constant.

Organizations achieve lasting success when they apply new technologies in ways that align with purpose, develop people, improve processes, and create value for customers. AI is no exception. When integrated thoughtfully into organizational systems, AI can accelerate learning, improve decision-making, strengthen performance, and support continuous improvement.

Ultimately, the impact of AI will be determined not by the technology itself but by the systems that guide its implementation and how well they are aligned to guiding principles. By understanding and adhering to the Shingo Guiding Principles, leaders can ensure that AI strengthens and improves systems, develops capability, and advances the pursuit of sustainable excellence by driving behavior closer to ideal, as informed by the principles.

Principles–Based Implementation

Like any tool, AI derives its value from the systems it enables and the behaviors those systems drive. The Shingo Guiding Principles provide a foundation for ensuring that AI implementation supports organizational purpose, develops people, improves systems, and creates value for customers. The following analysis explores the relationship between AI and each of the Shingo Guiding Principles and provides examples of ideal behaviors that reflect systems aligned to principles.

In developing ideal behaviors related to AI—and in particular those related to the ethical implementation of AI—the Shingo Institute borrows extensively from the following:

1. Oxford Institute for Ethics in AI; https://www.oxford-aiethics.ox.ac.uk
2. OECD AI Principles Overview; https://oecd.ai/en/ai-principles
3. ISO standard 42001; https://www.iso.org/standard/42001
4. AI tools, specifically MS Copilot and Box AI

Respect Every Individual

AI must be deployed to enhance human creativity, not to facilitate deskilling or replacement. By using AI to identify and prevent problems, the employee’s role as the primary improvement engine is protected. This follows Toyota’s practice of implementing technology “with a human touch.” Since technology lacks an inherent moral compass, it is necessary to be informed by this principle to implement AI in an ethical way. This respects not only every individual, but also humanity as a whole.

Examples of ideal behaviors include:

• Practitioners protect sensitive information when using AI.
• Practitioners verify AI-generated information before acting.
• Practitioners use AI to enhance thinking and learning.
• Practitioners continue developing personal expertise and judgment.
• Practitioners disclose significant use of AI.
• Practitioners exercise human judgment in all important decisions.
• Practitioners challenge AI outputs when evidence suggests otherwise.
• Practitioners use AI purposefully to solve problems and create value.
• Practitioners learn continuously about the capabilities and limitations of AI.
• Practitioners retain accountability for outcomes, regardless of AI involvement.
• The organization enhances the capabilities of people through the use of AI.

Lead with Humility

Leaders cannot delegate responsibility for AI to technology vendors or algorithms. They are accountable for strategy, culture, ethics, and performance. Humble leaders remain learners. They acknowledge the opportunities and limitations of AI, seek diverse perspectives, and encourage experimentation grounded in evidence. Leadership should focus on purpose, capability development, and creating psychological safety during change.

Examples of ideal behaviors include:

• Leaders use AI-generated insights to inform strategic decisions while retaining responsibility for setting organizational direction.
• Leaders align AI initiatives with organizational objectives and regularly review progress toward strategic goals.
• Leaders monitor and improve AI tools used in management systems to ensure they effectively support purpose, performance, and continuous improvement.
• Leaders model responsible AI use by applying it to improve their own effectiveness, learning, and decision-making.
• Leaders communicate in their own authentic voice and use AI to enhance—not replace—their thinking, judgment, and leadership.
• Leaders maintain accountability for decisions and outcomes and ensure that responsibility remains with people regardless of AI involvement.

Seek Perfection

This is the overarching Shingo Guiding Principle driving improvement. Broadly, organizations should implement AI because of the potential AI has to improve performance. The use of AI must be purposeful, aimed at improving safety/environment/morale, cost/productivity, delivery, quality, or customer satisfaction.

Purposeful use of AI can be achieved by combining AI with well-proven tools. Examples of ideal behaviors include:

• Standard Work
  Team members use AI to improve, follow, and continuously refine standard work based on process performance and learning.
• 5S
  Team members use AI to identify and eliminate workplace disorganization, making abnormalities easier to see and address.
• SMED
  Team members use AI to identify opportunities to reduce setup and changeover times while maintaining safety and quality.
• Value Stream Mapping
  Team members use AI to visualize end-to-end value streams, identify waste, and prioritize improvement opportunities.
• Visual Management
  Team members use AI to make performance, conditions, and abnormalities visible in real time so that action can be taken quickly.
• Total Productive Maintenance (TPM)
  Team members use AI to anticipate equipment issues, prevent unplanned downtime, and improve asset reliability.
• Statistical Quality Control (SQC)
  Team members use AI to analyze process variation, identify emerging risks, and support data-driven quality decisions.
• Problem Solving (PDCA, DMAIC, A3, etc.)
  Team members use AI to strengthen problem solving by generating insights, evaluating hypotheses, and supporting evidence-based decisions.

Embrace Scientific Thinking

Applying scientific thinking is the best way to make sure the implementation of AI is purpose driven. Every good experiment starts with identifying a specific business problem rather than starting with the technology itself. gThis ensures that the organization avoids the tendency to just throw AI—the tool—at a problem.

Examples of ideal behaviors include:

• Team members use AI to analyze customer needs and test improvements that increase the effectiveness of customer support.
• Team members use AI to evaluate candidate qualifications and improve hiring decisions using objective, evidence-based criteria.
• Team members use AI to test process changes, measure outcomes, and validate improvements before implementation.
• Practitioners use AI to compare relevant data, knowledge, and expert perspectives to improve the accuracy of diagnoses and decisions.
• Team members design experiments that test the accuracy and reliability of AI-generated outputs before applying them in practice.
• Team members regularly evaluate AI systems for bias, fairness, and completeness and use evidence to improve decision quality and outcomes.

Focus on Process

AI can improve visibility into work, helping teams understand process performance, identify waste, detect variation, and reveal opportunities for improvement. AI can also support standard work and informed problem solving rather than obscure decision-making behind black-box outputs.

Examples of ideal behaviors include:

• Team members use AI to analyze the flow of work and identify sources of waste, defects, delay, and variation within the process.
• Team members use AI to identify non-value-added activities and improve the effective use of time and resources.
• Team members use AI to analyze care delivery processes and identify opportunities to reduce delays and improve patient outcomes.
• Team members use AI to understand the customer or patient experience between process steps and identify opportunities to improve flow and continuity.
• Team members use AI to examine work from the perspective of the product, service, transaction, or customer and identify opportunities to improve process performance.

Assure Quality at the Source

AI can assist with identifying defects, errors, anomalies, and emerging risks. However, human oversight remains essential. Organizations should validate data quality, verify significant outputs, and establish controls proportionate to risk. Quality should be built into both operational processes and AI-enabled processes.

Examples of ideal behaviors include:

• Team members use AI to identify potential process failures, defects, and abnormalities and investigate their causes before problems reach the customer.
• Team members use AI-enabled detection systems to identify errors in real time and take immediate corrective action before defects occur.
• Practitioners use AI-supported analysis to evaluate relevant data, identify potential risks, and improve the accuracy and quality of decisions.
• Team members use AI to monitor process conditions, identify abnormal variation, and maintain process stability through timely intervention.
• Leaders apply the same quality standards, controls, and verification practices to AI-enabled processes as they do to operational processes.

Improve Flow and Pull

AI can help reduce delays, anticipate disruptions, improve scheduling, and increase responsiveness to customer demand. Application should be designed to improve flow, reduce waiting, and support pull-based systems rather than creating unnecessary complexity.

Examples of ideal behaviors include:

• Team members use AI to reduce administrative burden and increase the time available for value-creating work and customer service.
• Team members use AI to anticipate equipment issues and perform maintenance proactively to prevent disruptions to flow.
• Team members use AI to anticipate customer demand and align resources, inventory, and capacity with actual needs.
• Team members use AI to identify delays, bottlenecks, and waiting in workflows and take action to improve flow and responsiveness.

Create Constancy of Purpose

All AI implementations should be purposeful. By definition, this means AI improvements will lead the organization closer to fulfilling its clearly defined purpose. While AI can be a tool for maintaining constancy of purpose, it cannot create such constancy; rather, it can support leaders by providing information, tracking progress, and identifying trends. The purpose itself—and the empathy required to lead people toward that purpose—must come entirely from human leadership. AI can provide guardrails, data clarity, and mechanisms to ensure that purpose does not get lost in the daily noise.

Examples of ideal behaviors include:

• Leaders and team members use AI-generated insights to monitor performance and ensure daily actions remain aligned with the organization’s purpose and strategic objectives.
• Leaders use AI to identify activities, priorities, or resource allocations that may be drifting from organizational purpose and take corrective action when needed.
• Team members use AI intentionally and responsibly, ensuring its use supports organizational purpose and the effective stewardship of resources.

Think Systemically

AI can help teams understand interactions across functions, value streams, and systems. Decisions should evaluate both direct and indirect consequences, and improvements in one area should not create waste or instability elsewhere. Systems thinking requires balancing local optimization with enterprise-wide performance.

Examples of ideal behaviors include:

• Team members use AI to understand relationships across processes, functions, and value streams and evaluate the system-wide impact of decisions and changes.
• Team members use AI to test assumptions, evaluate potential consequences, and make decisions that optimize overall system performance rather than local results.
• Leaders use AI to evaluate innovation opportunities by considering their impact on customers, people, processes, and the overall enterprise system.

Create Value for the Customer

Every significant AI application should create measurable value. Benefits include improvements to safety/environment/morale, cost/productivity, delivery, quality, and customer satisfaction. If value cannot be identified, the initiative should be reconsidered.

Examples of ideal behaviors include:

• Team members use AI to improve quality, responsiveness, and reliability in ways that create measurable value for customers.
• Team members evaluate AI initiatives based on their ability to improve customer outcomes and reduce waste.
• Team members use AI to better understand customer needs and improve the products and services they provide.
• Team members use AI to identify and eliminate defects, delays, and barriers that negatively impact the customer experience.
• Leaders prioritize AI investments that demonstrably improve customer value, safety, quality, or service.
• Team members measure the impact of AI-enabled improvements through customer-focused results and feedback.
• Leaders and team members discontinue or redesign AI applications that do not create meaningful customer value.
• Team members use AI to anticipate customer needs and proactively improve customer experience.
• Team members use AI to increase the value delivered to customers while reducing effort, delay, and complexity.
• Leaders ensure AI initiatives remain aligned with customer needs, organizational purpose, and long-term value creation.

AI and Systems

Tools alone do not create sustainable improvement—systems do. Lasting results occur when new capabilities are intentionally integrated into work systems, management systems, and improvement systems. As with any technology, the impact of AI will be limited unless it becomes part of the systems that guide daily work, decision-making, and continuous improvement.

When implemented purposefully, AI has the potential to strengthen virtually every organizational system by improving learning, visibility, decision quality, and the flow of information.

Key Points

• AI can support all organizational systems, including:
• Work Systems that deliver products and services to customers
• Improvement Systems that develop safety, capability, problem-solving, continuous improvement, etc.
• Management Systems that align strategy, deploy objectives, monitor performance, and guide decision-making
• Responsibility for the use of AI should be clearly defined within each system. The system owner remains accountable for system performance regardless of the level of AI involvement.
• AI can enhance many of the tools that support effective systems, including standard work, reports, scheduling, visual management, feedback mechanisms, problem-solving processes, improvement tracking, etc.
• Every system should be evaluated using meaningful measures of both behaviors and results. AI can help identify patterns, analyze performance, and support the development of meaningful Key Behavioral Indicators (KBIs) and Key Performance Indicators (KPIs).
• AI specialists and technologists should work alongside operational teams (e.g., industrial engineers or process-improvement specialists) to understand customer value, workflows, system interactions, sources of waste, and opportunities for improvement.
• Organizations should recognize that developing AI capability requires learning and experimentation. As with any new skill, a temporary reduction in productivity may occur while individuals and teams learn how to apply AI effectively.
• AI systems should be continuously improved. Organizations should regularly evaluate AI applications to ensure they remain effective, accurate, aligned with organizational purpose, and supportive of people and performance.

Human Responsibility Within Systems

AI can support system performance, but it cannot replace human ownership of systems. Leaders and team members remain responsible for defining problems, making decisions, implementing improvements, and achieving results.

AI can assist by:

• Generating ideas, alternatives, and hypotheses for teams to evaluate.
• Identifying anomalies in standard work and operational performance.
• Analyzing data and generating reports to support learning and decision-making.
• Providing real-time feedback on system performance.
• Monitoring system conditions and highlighting emerging risks.
• Tracking activities, commitments, and schedules to support execution and accountability.

Importantly, responsibility for the performance of the system always remains with people.

Knowledge Sharing Across Systems

One of AI’s greatest opportunities may be its ability to accelerate organizational learning. AI can help capture, organize, and distribute knowledge across teams, facilities, functions, and enterprises. Improvements developed in one location can be rapidly shared, understood, adapted, and applied elsewhere, increasing the rate of learning throughout the organization.

Continuous Reflection

As AI capabilities continue to evolve, organizations should regularly review their systems and ask:

• How can AI help us better achieve the purpose of this system?
• How can AI strengthen the performance of this system?
• How can AI improve the effectiveness of the tools within this system?
• How can AI help develop the capabilities of the people who operate this system?
• How can AI help us create greater value for customers?

These questions help ensure that AI remains a means of improving systems and developing people rather than becoming an objective in itself.

Conclusion

AI has the potential to accelerate learning, improve decision-making, strengthen systems, and enhance an organization’s ability to create value. However, technology alone does not create excellence; sustainable results emerge from the interaction of principles, systems, tools, results, and culture.

The world’s best organizations achieve sustainable excellence not because they adopt new technologies but because they align purpose, people, processes, and systems to create value for customers. Technology serves as an accelerator of these capabilities, not a substitute for them.

The same is true for AI. It can strengthen work systems, management systems, and improvement systems by refining quality, flow, productivity, learning, and innovation. However, purpose, culture, judgment, and accountability remain the responsibility of leaders and the people they lead.

Ultimately, the impact of AI will be determined by the systems that guide its use and how well they are aligned to guiding principles. Organizations that apply AI to develop people, improve systems, and create greater value for customers are more likely to achieve sustainable excellence than those that pursue technology for its own sake.

By anchoring AI implementation to the Shingo Guiding Principles, organizations can ensure that technological advancement remains aligned with organizational purpose, respect for people, and the pursuit of sustainable excellence.

Fundamentally, the future of AI is not a technology question—it is a leadership question. Organizations that understand and adhere to enduring principles as they adopt emerging technologies will be positioned to achieve sustainable excellence.  

[1] Shingo Model (https://shingo.org/model.)