Why Traditional Organizations Need a Different Approach to AI Readiness

Key Takeaways

• AI readiness is an organizational capability, not simply employee access to AI tools.
• Traditional organizations have to introduce AI into established workflows, legacy technology, management structures and governance requirements.
• Redesigning work around AI is more important than simply increasing the number of employees using AI.
• Managers are central to AI readiness because they influence experimentation, performance expectations and whether workflows can actually change.
• Role-specific training, controlled experimentation and decentralized use-case discovery can help organizations move from adoption to measurable value.
• Traditional organizations also have an advantage: years of institutional knowledge that AI can potentially make easier to access and reuse

AI adoption is moving quickly. Organizational change is not always keeping pace. That gap is becoming increasingly visible.

According to McKinsey’s 2026 research on AI transformation, 70% of respondents said they felt personally prepared to adopt and use AI. Yet only 27% of leaders believed their organizations were ready to make the changes required for an AI-enabled future.

The same research found that organizational readiness accounted for 48% of the difference between leaders who reported capturing value from AI and those who did not. Personal readiness accounted for 25%.

In other words, your employees can become comfortable with AI faster than your organization can change around them.

And for traditional organizations like banks, manufacturing companies, retailers, logistics business, healthcare providers, or a construction company, that poses a particular challenge.

The reason is these organizations already have workflows, hierarchies, technology systems, risk controls, job structures and deeply established ways of working.

AI readiness in such environments requires more than giving employees access to new tools.

It requires changing the system around those tools.

Why Are Traditional Organizations a Different AI-Readiness Case?

A digitally native company may be able to design a new process around AI from the beginning.

An established business usually has to ask a harder question:

What happens to everything that is already there?

Existing workflows have years of history behind them

Processes in traditional organizations often developed over years or decades.

An insurance claim may pass between several departments. A manufacturing decision may depend on engineering, maintenance, procurement and safety teams. Recruitment may require multiple approvals before a vacancy is even advertised.

If you introduce AI without examining these workflows, you can end up with a predictable outcome:

Your organization performs the same process slightly faster.

McKinsey’s research suggests that this is not where the greatest value lies. Among organizations at the early “enablement” stage of AI adoption, leaders were 5.3 times more likely to report enterprise value when workflows had been redesigned than when they remained unchanged.

Legacy technology and fragmented data can limit what AI can do

Your organization may operate across ERP systems, spreadsheets, internal databases, proprietary applications, old APIs and departmental data silos.

The problem is not always a lack of data.

You may already have plenty of it.

The difficulty is making the right information accessible, reliable and secure enough for AI systems to use.

The Cisco AI Readiness Index 2025 identified a relatively small group of organizations — about 13% of those surveyed — as “Pacesetters.” These companies were four times more likely to move AI pilots into production and 50% more likely to report measurable AI value.

Cisco’s findings also highlight the importance of the infrastructure, data, governance and talent foundations surrounding AI.

This is one reason you cannot treat AI readiness as an issue for employees or HR alone. Technology architecture, data governance and workforce readiness have to develop together.

Traditional organizations often operate with higher consequences for failure

An AI mistake in a brainstorming session is very different from an AI mistake affecting lending, medical care, industrial safety, regulatory reporting or customer data.

If you operate in a highly regulated or operationally sensitive industry, caution is understandable.

But caution does not have to mean avoiding experimentation.
The better question is:

How can you create environments where employees can experiment without removing the controls your organization needs?

Established organizations also have something valuable: institutional knowledge

Legacy is not only a disadvantage.

Your organization may possess years of customer history, technical expertise, operating procedures, policies, training materials and knowledge held by experienced employees.

AI may make some of that accumulated knowledge easier to distribute.

In the NBER study Generative AI at Work, researchers studied the introduction of a generative AI assistant among 5,179 customer-support agents. Access to the tool increased productivity by 14% on average, with a 34% improvement among novice and lower-skilled workers.

The researchers also found suggestive evidence that the AI system helped disseminate some of the practices of more capable workers to less experienced colleagues.

For an organization that has operated for decades, that possibility is significant.

AI Usage Is Not the Same as AI Readiness

AI adoption numbers can create the impression that organizations are transforming rapidly.

The reality is more complicated.

The UK Office for National Statistics’ 2026 analysis of AI use in businesses found that the proportion of businesses with at least 10 employees reporting use of an AI technology had increased from around 12% in late 2023 to around 35% by June 2026.

But the ONS described adoption as relatively shallow. The average number of AI technologies used per adopting business had increased only modestly, from around 1.4 to 1.6.

Adoption can therefore become broad without becoming deep.

You can buy AI licences, introduce a chatbot, run prompting workshops and publish an AI policy without fundamentally changing how work gets done.

This is why you should distinguish between two questions:

Are our employees using AI?

and

Can our organization consistently convert AI into better work?

The second question is closer to genuine AI readiness.

For HR leaders in particular, understanding what AI readiness actually means for HR teams requires looking beyond individual tools towards skills, roles, workflows, governance and workforce planning.

Start With Workflows, Not AI Tools

Instead of beginning with:

“Where can we use ChatGPT?”

start with:

“How does this work actually happen today?”

Take recruitment as an example:

Vacancy approval → role definition → sourcing → screening → interviews → references → offer → onboarding.

At every stage, you can ask:

• Where does human judgment matter most?
• Where could AI assist an employee?
• What could be automated safely?
• What information would the AI system require?
• Where should human review remain compulsory?
• Could the process be designed differently altogether?

You can apply the same exercise to procurement, maintenance, claims, customer service, sales, finance, logistics and compliance.

Recent INSEAD research into what the authors call the AI “mapping problem” illustrates why this matters.

In a field experiment involving 515 high-growth startups, firms exposed to examples of how other businesses had reorganized production around AI identified 44% more AI use cases than the control group.

The study involved startups rather than traditional enterprises, so its exact results should not simply be transferred to every established business.

But the underlying challenge is highly relevant: before you can scale AI, you first have to discover where and how it creates value within the way your organization produces work.

The challenge is partly technological. It is also a mapping problem.

Give Employees Safe Places to Experiment

Your employees will not necessarily wait for an official transformation programme before trying AI.

That creates both an opportunity and a governance problem.

BCG’s 2025 AI at Work research, based on responses from more than 10,600 workers across 11 countries, found that when employees do not have the AI tools they believe they need, more than half say they will find alternatives and use them anyway.

If your response is simply to restrict access, you may create shadow AI use rather than stop experimentation.

A more practical approach is to create clear boundaries around what employees can do.

That can include:

• approved AI tools;
• secure internal environments;
• rules for confidential and customer data;
• AI sandboxes;
• mandatory human review for high-risk outputs;
• escalation procedures;
• clear accountability for AI-assisted decisions.

DBS provides one example of how this can work inside a heavily regulated organization.

According to the bank’s 2025 annual reporting on its AI transformation, DBS had deployed more than 2,000 AI models across more than 430 use cases, generating approximately SGD 1 billion in economic value from data analytics and AI/ML initiatives.

Its technology strategy also included a GenAI framework covering reusable components, governance guardrails, workflow capabilities and prompt engineering.

You do not have to choose between control and experimentation.

Your organization needs both.

This is especially relevant in financial services, where AI’s impact on banking jobs across the UAE and GCC is already forcing employers to think about both technology adoption and how roles may evolve.

Train Employees for Their Work, Not Simply for AI

A generic AI awareness session can help your employees understand what the technology is.

It will rarely be enough to change how they work.

BCG’s AI at Work 2025 findings show that regular AI usage was considerably higher among employees who had received at least five hours of training, particularly when training included in-person learning and coaching.

Your training strategy should therefore reflect how different groups actually interact with AI.

• Executives need to understand business value, limitations and operating-model implications.
• Managers need to learn how to redesign workflows, evaluate AI-assisted output and manage teams whose roles are changing.
• Knowledge workers may require deeper skills around research, verification, analysis and workflow automation.
• Frontline employees may need task-specific tools, mobile interfaces, voice capabilities and practical microlearning.
• Technical teams need capabilities in integration, evaluation, architecture and monitoring.
• Risk, legal and compliance teams need to understand privacy, auditability, accountability and model risk.

DBS, for example, has introduced role-based AI learning for different technology roles, combining technical capabilities with business understanding, change management and workflow automation.

For organizations operating in the Emirates, this also makes building AI readiness for HR teams in the UAE a workforce-design issue rather than simply a technology-training exercise.

Middle Managers Can Make or Break AI Readiness

You may have executive approval for an AI strategy.

Your employees may be eager to experiment.

But managers often determine what happens between those two levels.

They influence whether employees have time to learn, whether experimentation is encouraged, whether mistakes are tolerated and whether an old process is actually allowed to change.

They can also experience uncertainty themselves.

McKinsey found that one in four middle managers surveyed expressed anxiety about AI-related changes, compared with one in five individual contributors.

This should not be interpreted as evidence that managers are obstacles.

It means you need to make them participants in the transformation.

The Microsoft 2026 Work Trend Index reached a similar conclusion about the importance of the environment around employees.

Microsoft found that organizational factors — including culture, manager support and talent practices — accounted for 67% of reported AI impact in its analysis, compared with 32% for individual mindset and behaviour. Microsoft describes this as an association rather than proof of causation, but the pattern is clear: individual capability alone is not enough.

If you want AI adoption to change how work gets done, AI capability also has to become a management capability.

Managers should be able to show what their teams have tested, what has changed and what they have learned.

Find the AI Users Who Already Exist Inside Your Company

Before you launch another company-wide AI initiative, find out what employees are already doing.

Ask:

• Which AI tools are you already using?
• What are you using them for?
• Where are they helping?
• Where are they creating problems?
• Which unofficial workflows have already appeared?
• What would make employees more comfortable using approved tools?

The answers may reveal use cases that a central transformation team would never identify on its own.

Some of these employees can also become AI champions who help colleagues experiment safely.

Useful initiatives can include:

• AI office hours;
• departmental AI champions;
• internal demonstrations;
• communities of practice;
• workflow libraries;
• reusable prompt or automation repositories;
• internal hack days;
• sessions where teams explain what failed as well as what worked.

However, be careful not to create a small AI elite.

The role of champions should be to spread capability rather than concentrate it.

Combine Central Governance With Local Experimentation

You do not need every department building its own AI strategy independently.

But you also do not need every use case dictated from headquarters.

A hub-and-spoke model can provide a middle ground.

Your central AI function can establish:

• approved platforms;
• data and security standards;
• governance;
• technical architecture;
• evaluation frameworks.

Business units can then focus on:

• identifying use cases;
• understanding local workflows;
• testing AI;
• measuring results;
• adapting solutions to domain needs.

HR and L&D can address skills, job redesign, workforce planning and capability building.

Legal, risk and compliance teams can define additional controls for sensitive applications.

This allows you to combine knowledge held close to the work with the governance required to operate safely.

Decide What Happens to the Time AI Saves

One of the most overlooked AI culture questions comes after productivity improves.

What happens to the time your employees save?

Do you give them more work?

Use the capacity to reduce costs?

Invest it in training?

Create more customer-facing time?

Allow further experimentation?

The IBM Institute for Business Value’s 2026 CHRO Study found that only 42% of organizations primarily direct AI productivity gains towards innovation or reskilling. Others split those gains between reinvestment and savings, capture them as cost reduction or profit, or lack a clear approach.

What you do with those gains sends employees an important signal.

If every productivity improvement simply results in a larger workload, employees may have little incentive to surface how much time AI is actually saving them.

If you reinvest some of that capacity in learning, innovation, better customer service or career development, AI can begin to support a broader learning culture.

This is also where HR consulting can support organizations by connecting AI adoption with job design, workforce planning, manager capability and the wider people strategy rather than treating AI as an isolated technology initiative.

Measure Business Change, Not AI Activity

Licence counts are easy to measure.

So are prompts, chatbot logins and training attendance.

None of them proves that AI is improving your organization.

Look at outcomes such as:

• cycle-time reduction;
• quality improvements;
• error reduction;
• customer outcomes;
• employee productivity;
• cost savings;
• time to competency;
• workflows redesigned;
• pilots moved into production;
• employee experience;
• new revenue or service capabilities.

You should also capture lessons from experiments that fail.

A failed pilot can still create value if you understand why it failed and make that knowledge available to the next team.

In Closing…

Traditional organizations are often described mainly in terms of what holds them back.

Legacy technology, slow decision-making, complex governance. And established ways of working.

But if you lead an established organization, you may also have assets that younger competitors have spent years trying to acquire: customer relationships, proprietary information, operating history, domain expertise and employees with decades of experience.

AI can potentially make those assets more accessible and reusable.

You do not need to turn your organization into an AI startup.

The more important goal is to combine what your organization already knows with a faster way of learning.

You also need employees who understand how to use the technology, managers who make experimentation possible, systems that provide reliable information, governance that establishes safe boundaries and leaders willing to redesign work when the evidence points to a better way.

That is what makes an organization AI-ready.

Frequently Asked Questions

What does it mean for an organization to be AI-ready?

Being AI-ready means having the technology, skills, workflows, governance and management practices needed to integrate AI safely into everyday work and turn its use into measurable organizational value.

Why is AI adoption more difficult for traditional organizations?

If you operate an established organization, AI has to work around existing workflows, technology, organizational structures and governance requirements. You are integrating it into an operating system that already exists rather than designing everything from scratch.

How can traditional companies build an AI-ready culture?

Start by mapping how work happens today, provide employees with safe tools to experiment, introduce role-specific training, involve managers, develop internal AI champions and establish governance that supports useful experimentation.

What role should HR play in AI readiness?

HR can help you identify changing skills requirements, redesign roles, prepare managers, support workforce planning, develop AI learning programmes and communicate clearly about how AI-related changes may affect employees.

How should companies measure AI readiness?

Look beyond AI usage. Measure outcomes such as workflow redesign, employee capability, quality, productivity, customer outcomes, safe adoption and your ability to turn successful experiments into repeatable business processes.