EPISODE 03: Your Organisation Isn’t AI-Ready Just Because Your People Attended AI Training

Posted on September 9, 2026 by Rowell Mariano

You probably remember one idea I kept coming back to if you happen to read Episode 02 of The Backstage: Beyond the Classroom. Learning excellence is learning translated into actual performance. We can train people, give them knowledge, help them practice a skill, and even assess what they learned. But at some point, we still have to ask if they can actually use that learning when it matters. So this week, I want to take that same thinking into a conversation almost every organisation seems to be having right now. Let us talk about artificial intelligence.

Imagine your CEO tells you, “We have already trained 80% of our employees on AI. Are we AI-ready now?” My answer would be simple. No. You may have created awareness. You may even have developed some level of AI capability among your people. But organisational AI readiness requires much more than the percentage of employees who completed an AI course.

That distinction matters because organisations can easily confuse learning activity with organisational readiness. You can train thousands of employees on prompt writing. You can run webinars on generative AI. You can give people certificates and access to several platforms. But if nobody is clear about the problems AI is supposed to solve, the work remains exactly the same, leaders have not established clear rules, and employees are unsure about what they can safely use, then the organisation is still trying to figure things out.

Interestingly, recent research shows that this gap is already happening. In a 2026 McKinsey study, 70% of respondents said they felt personally prepared to adopt and use AI, yet only 27% of leaders believed their organisations were ready to make the changes needed for an agentic future. The same research found that organisational readiness was much more strongly associated with enterprise value than personal readiness alone (De Smet et al., 2026). In other words, your employees may be ready before your organisation is.

AI Has Been Around Longer Than the Current Hype

I remember being invited to speak to an association back in 2017 about emerging trends in learning. I talked about artificial intelligence and the growing use of technology in training. I even talked about the use of cameras in delivering learning. Some people laughed and giggled. The idea of sitting in front of a camera to conduct training sounded strange to some people at the time, and AI still felt like something you would encounter in science fiction.

By 2018, I was already speaking about artificial intelligence through the Philippine Society for Talent Development. I also discussed technologies such as augmented reality, virtual reality, and mixed reality. These were conversations happening years before generative AI became part of everyday workplace discussions. I am sharing this because organisations sometimes become interested in change only when the change becomes impossible to ignore. By then, everyone is rushing to catch up.

Artificial intelligence itself is not new. John McCarthy is widely credited with introducing the term in the 1955 proposal for the Dartmouth Summer Research Project on Artificial Intelligence. The actual research project took place in 1956 and became an important milestone in the development of AI as a formal field (McCarthy et al., 1955).

What has changed is what AI can now do and how accessible it has become. Generative AI can create new content such as text, images, audio, and other digital outputs based on patterns learned from data (National Institute of Standards and Technology, n.d.). Agentic AI moves further because AI agents can work toward goals, coordinate steps, interact with tools, and perform parts of a process with greater autonomy. Deloitte’s 2026 research found that many leaders already expect AI agents to significantly reshape jobs and business processes in the coming years (Deloitte, 2026).

The technology will continue to move. The question for organisations is not simply how quickly they can follow the latest AI trend. The better question is how prepared they are to make good decisions about where AI belongs in the organisation and what needs to change because of it.

Do Not Start With the AI Tool

One of the mistakes organisations can make is starting with the technology instead of starting with the problem. Someone hears about a new AI platform and the immediate conversation becomes, “Should we buy it?” Another organisation sees a competitor using AI and suddenly wants an enterprise-wide rollout. Then training follows because everybody needs to know how to use the new tool.

I would start somewhere else. What organisational problem are we trying to solve, improve, assist, or augment? That question should come before the technology purchase and before the training calendar.

Look at the work itself. Where are people spending too much time? Which processes have unnecessary steps? Which activities are highly repetitive? Where are errors happening? Which decisions could benefit from better information? Where can AI help people complete work faster without compromising quality? Once the organisation has identified those points, the conversation about technology becomes much more useful because you already know what the technology is expected to accomplish.

This thinking is also consistent with current research on organisational AI readiness. Jonak and Wodecki (2026) argue that AI readiness is becoming less about simply possessing technology and more about the organisation’s ability to coordinate governance, processes, implementation, and employee behavior so AI can produce reliable outcomes. Their research treats readiness as an evolving organisational capability rather than a static inventory of tools and resources.

That is also why I would be cautious when an organisation immediately says, “We want AI across the entire enterprise.” Start with critical areas. Study the current process. Identify the use case. Establish the baseline. Then you can decide how far the solution should go.

Training and Technology Need to Move Together

Once you know where AI will be used, the tools and capability building should move in parallel. It makes little sense to train employees extensively on technology they cannot access afterward. It makes just as little sense to spend heavily on AI platforms and assume employees will know how to use them responsibly and effectively.

This is where I think some AI initiatives become disconnected. The technology team may be purchasing systems while the learning team is conducting training, but the actual employee experiences them as two separate things. People attend an AI course and become excited about what is possible, only to return to work and discover that they do not know which tools are approved. They are unsure what information they are allowed to enter. Their manager cannot explain where AI fits into the process. Eventually, people either stop using what they learned or begin experimenting without enough guidance.

Technology should be available when people are developing the capability to use it. At the same time, employees need enough knowledge to understand what the technology can and cannot do. They need practice using it within the context of their actual responsibilities. If the organisation wants people to adopt AI, employees need both capability and access.

This is exactly why attending AI training alone does not make an organisation AI-ready. Training develops people. Organisational readiness requires the organisation around those people to be ready too.

If the Work Did Not Change, What Exactly Did AI Change?

Suppose an employee used to spend four hours every week preparing a report manually. The organisation introduces AI and the employee can now complete much of that work in two hours. That sounds good. But then management still asks the employee to complete the entire old manual process as well because “that is how we have always done it.”

What happened? The organisation introduced AI and created more work.

This is why AI adoption should eventually lead to process and job redesign. If a repetitive activity can now be automated or completed much faster with AI, organisations need to ask what happens to the work that remains. Which parts should be automated? Which activities should be performed by a person working with AI? Which responsibilities still need to remain primarily human because they require professional judgment, accountability, relationships, or context that cannot simply be handed to a system?

McKinsey’s 2026 research found that leaders were 5.3 times more likely to report enterprise value when workflows had been redesigned around AI compared with organisations where workflows remained unchanged. The same research found that organisations further along in AI transformation were changing roles, workflows, operating practices, and behaviors instead of simply giving employees access to AI tools (De Smet et al., 2026).

Deloitte reported a similar challenge with agentic AI. Only 5% of organisations in its 2026 survey said their business processes were highly prepared for AI agents. Many organisations were still adding agents on top of existing processes rather than redesigning the process itself (Deloitte, 2026).

So if AI saves your employees time but their jobs remain exactly the same, the work is not finished. The goal should not be to give people another system to manage. The goal should be to improve how work gets done.

People Still Matter, But Their Work Will Change

The conversation about AI eventually reaches the same question. Will AI replace people?

I would not give anyone a comforting absolute answer because parts of work are already changing. Some repetitive activities will be automated. Certain jobs will look very different. New roles will emerge, and some existing responsibilities may disappear. Deloitte’s 2026 study found that 43% of surveyed leaders expected agentic AI to significantly disrupt their workforce within the next 12 to 18 months, including changes in job requirements and the creation of new ways of working (Deloitte, 2026).

For me, the more useful conversation is about adaptation. If much of your work is repetitive and AI can perform parts of it faster, then your value cannot remain dependent on doing the same repetitive work manually. You need to ask how your expertise works with AI and what higher-value contribution you can make because the technology now handles some of the tasks that used to consume your time.

But organisations cannot put all the responsibility on employees. If a role changes because of AI, leaders need to help redesign that role. People may need new capabilities. Expectations may need to change. Managers also need to adapt how they lead because they cannot tell employees to work differently while measuring them using the same old standards.

There will also be fear. Some employees will worry about what AI means for their jobs, and telling them simply not to worry will not solve that. Leaders need to communicate what is changing, what they already know, and what they are still trying to determine. McKinsey’s research found that trust in the organisation was an important factor in AI readiness, especially when employees were dealing with uncertainty about how their work might change (De Smet et al., 2026).

Access to AI Does Not Mean Unrestricted Use

Another part of AI readiness is responsible use. Employees need to know what they can use, where they can use it, and what they should never place into an AI system.

If your organisation is using public AI tools, employees should not assume that they can enter confidential company information, personal information, proprietary material, or sensitive client data simply because the tool is easy to access. The organisation needs policies and processes that clearly define what is approved and what is restricted.

The same applies to AI-generated output. AI can assist the work, but it does not remove accountability for the work. If an employee uses AI to create a report, recommendation, learning material, client communication, or analysis, that person still needs to review and curate what will actually be used. Copying and pasting AI output without checking accuracy, context, bias, or suitability creates unnecessary risk.

Recent research reinforces the growing importance of governance. Jonak and Wodecki (2026) found that governance and accountability have become major parts of contemporary AI readiness as the technology becomes easier to access. Deloitte’s 2026 research also found that risk, security, governance, and business process preparedness continued to lag behind organisations’ ambitions for agentic AI (Deloitte, 2026).

Being AI-ready therefore does not mean allowing everyone to use everything. It means giving people useful access within clear boundaries and making sure accountability remains clear.

So What Would an AI-Ready Organisation Look Like?

If an organisation asked me to assess its AI readiness, I would not begin by asking how many employees completed AI training. I would look at a much wider set of conditions. To make that practical, I would organise those conditions through what I call the AI READY Framework.

A: Align Leadership on the Problem and Intended Value

AI readiness should begin with leadership being clear about what the organisation is trying to accomplish. Where should AI improve performance? What problem needs attention? What type of work could benefit most? Leaders do not need to become AI engineers, but they do need enough understanding to make informed decisions about the direction of the organisation. This is also where leadership needs to establish the reason behind the change so employees are not simply being told to “use AI” without knowing what success should look like.

I: Identify Use Cases and Establish the Baseline

Once the problem is clear, identify the specific processes and tasks where AI can help. Look at current performance before making changes. How long does the process take today? How much repetitive effort is involved? What errors occur? What is the current quality of the output? Without a baseline, it becomes difficult to prove later that AI actually improved anything.

R: Rules, Governance, and Responsible Use

Establish the rules before AI becomes deeply integrated into the work. Define approved tools, access levels, data restrictions, review requirements, decision accountability, and other controls needed for responsible use. Governance should fit the type of AI being used because an employee asking generative AI to help draft content carries a different level of risk from an autonomous agent performing several steps inside a business process.

E: Equip People and Technology in Parallel

Give employees access to the right technology while developing their ability to use it. AI capability is not simply knowing how to write a good prompt. People need to know where AI fits into their work, how to check outputs, when not to use it, and how to combine their professional expertise with what the technology can provide. Capability building should continue as the technology changes.

A: Adapt Work, Workflows, and Roles

Once AI becomes part of the process, examine the job itself. Reduce appropriate repetitive work. Automate tasks that make sense to automate. Identify areas where AI should assist rather than replace the person doing the work. Then look at what employees can spend more time doing because some administrative burden has been reduced. This is where productivity should translate into better work rather than simply asking people to complete more tasks in the same amount of time.

D: Drive Adoption Through Leadership Follow-Through

Leadership comes back into the process because introducing AI is not enough. Managers need to support people as they adjust, reinforce the new process, respond to concerns, and make sure employees are not being pushed back into the old way of working. If leaders say they want AI adoption but continue rewarding the same inefficient behaviors, people will eventually return to what is familiar.

Y: Yield Measurable Value and Keep Adjusting

Finally, look at what AI produced for the organisation. Compare the results with the baseline. Did turnaround time improve? Did repetitive work decrease? Is output quality better? Are errors being reduced? Are employees becoming more productive? Is the customer experience improving? Has the organisation gained measurable financial or operational value? AI readiness is not something you assess once and forget because technology and organisational needs continue to change. Jonak and Wodecki (2026) similarly argue that modern AI readiness should be treated as an evolving organisational capability rather than a fixed state.

The AI READY Framework therefore needs to operate as an iterative process. Once results begin to plateau, the organisation should review what has changed. A technology may have improved. A task may no longer need to exist. A new risk may have emerged. Another capability may now be required. The organisation goes back, studies the work again, and adjusts.

AI Readiness Is Bigger Than AI Training

Training still matters. In fact, capability building is one of the conditions that will determine how successfully people use AI. But training should sit inside a much bigger organisational effort. We should stop assuming that sending people to AI workshops automatically prepares the organisation for what comes next.

Your people may know how to use AI while your processes are still built for the old way of working. They may understand generative AI while your policies remain unclear. They may have access to powerful tools while managers continue to expect the same manual work. They may even be ready to adapt while the organisation itself has not decided where AI should create value.

That is the real readiness gap.

The latest research is telling us something very similar. Organisations getting better results are not simply buying more technology or encouraging more employees to experiment. They are changing work, governance, leadership practices, and organisational behavior around what AI now makes possible (De Smet et al., 2026; Jonak & Wodecki, 2026).

So the next time someone tells you that your organisation is AI-ready because hundreds or thousands of employees have completed AI training, ask a few more questions. What problems are you solving? How has the work changed? What are people now able to do differently? What value has the organisation actually gained?

If your organisation is asking how to move beyond AI training and build real AI readiness, this is exactly the kind of capability and organisational work I help leaders examine. Because the real question is not simply how much your people know about AI. It is how ready your organisation is to work differently because of it.

That is another part of the work that happens in the backstage, beyond the classroom.

References

De Smet, A., Goldstein, D., Price, H., & Catlin, T. (2026, July 8). From adoption to impact: Three horizons of AI transformation. McKinsey & Company. https://www.mckinsey.com/capabilities/people-and-organization/our-insights/from-adoption-to-impact-three-horizons-of-ai-transformation?utm_source=Horizonscan&utm_medium=email&utm_campaign=ai-pulse-daily-brief-2026-08-03

Deloitte. (2026, August 12). AI agents are only the beginning: Deloitte survey examines the AI readiness gap and reveals how enterprises can prepare for agentic success. https://www.deloitte.com/us/en/about/press-room/deloitte-survey-examines-ai-readiness-agentic-ai-success.html

Jonak, K., & Wodecki, A. (2026). The evolution of organizational AI readiness toward an orchestration capability. Discover Artificial Intelligence, 6, Article 912. https://doi.org/10.1007/s44163-026-02019-z

McCarthy, J., Minsky, M., Rochester, N., & Shannon, C. (1955, August 31). A proposal for the Dartmouth Summer Research Project on Artificial Intelligence. Dartmouth College. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/http://jmc.stanford.edu/articles/dartmouth/dartmouth.pdf

National Institute of Standards and Technology. (n.d.). Generative artificial intelligence. Computer Security Resource Center Glossary. https://csrc.nist.gov/glossary/term/generative_artificial_intelligence