Hesham Khalafallah
In my two previous articles, I wrote about moving AI from recommendations to execution, and about five use cases an organization can start with. But there is a question more important than the tool itself: who is going to use it? And are they ready?
Three waves I lived through, and the same problem
I clearly remember the years after 2005, when many organizations decided to bring computers into most of their departments and operations, and learning basic computer skills became a necessity for employees. I was working in training at the time, and I saw first-hand how many employees struggled to learn, and how some stood against the idea altogether.
The same situation repeated itself in 2011 and 2012, when talk of digital government, e-services and reliance on the internet began. And it happened a third time with the COVID-19 crisis, when distance learning and e-learning became an imposed reality overnight.
In all three cases, the real problem, in my view, was the same: digital solutions were imposed on employees without any psychological preparation at the start. Employees found themselves facing a new system they had not been consulted about, with no explanation of why it had come or how it would help them. So they resisted it, or used it reluctantly and as little as possible.
Today, with artificial intelligence, we are facing the fourth wave. The question is: will we repeat the same mistake?
Step one: introduce before you implement
Before any tool or system, an organization should begin with simple introductory seminars and meetings about AI: what it is, what its advantages and limits are, and how exactly it can help employees in their daily work. The goal here is not technical training, but building understanding and dispelling fear.
The best way in is to show employees that they already use AI in their daily lives, directly or indirectly: in map apps, instant translation, the voice assistant on their phone, and the content suggestions in the apps they follow. The technology is not as foreign to them as they think; it is already part of their day.
Step two: needs start with the employees themselves
Once employees understand the technology, we ask them to make the proposals: which tasks take up your time? What would you like AI to help you with? Solutions are then offered based on these proposals.
This is where the big difference happens. Employees will accept the solution, and even welcome it, because they are convinced it meets their needs and requests, and because they have become part of the development itself, not just its recipients.
Psychological challenges: their causes and how to overcome them
• Fear for one’s job: caused by the popular image of “the machine that replaces people.” We overcome it with clarity: the aim is to eliminate repetitive tasks, not employees, and to highlight examples of colleagues whose work became easier and more valuable.
• Fear of failing or looking incompetent: especially among older employees or those with long experience. We overcome it with a safe training environment, small groups, and encouraging experimentation without blame.
• Resistance to change: because the old way is familiar and comfortable. We overcome it by involving employees in the decision from the start, as described above.
• Distrust of the results: especially after hearing about AI mistakes. We overcome it by stressing that a human stays in the loop and has the final review.
Technical challenges: their causes and how to overcome them
• Weak basic digital skills: addressed through foundation training before the AI tools themselves.
• Legacy systems and unorganized data: addressed by preparing the digital environment gradually, and organizing the data first in the department where implementation will start.
• Security and privacy concerns: addressed through a clear internal policy defining what can and cannot be entered into the tools.
• The language challenge: some tools are weaker in Arabic. We address this by choosing tools with good Arabic support and testing them on real examples from the organization’s own work before adopting them.
Phased implementation: three tracks in parallel
One of the most important lessons from previous waves is that implementation must be phased, with three tracks moving forward together at the same time:
• Application: one use case after another, one department after another.
• Training and qualification: preceding and accompanying each phase, not coming after it.
• The digital environment: devices, systems, data and policies, prepared step by step.
When these tracks move in parallel, the whole system falls into place and becomes organized, and implementation is completed without shocks.
Programs that suit each group
No single training program fits everyone. Senior leaders need programs focused on decision-making, governance and measuring impact; middle management needs programs on change management and following up on implementation; and front-line employees need hands-on training on the tools in their actual tasks. Differences in age and experience must also be taken into account: what suits a young employee may not suit a colleague with thirty years of service, and vice versa.
Conclusion
The success of AI in any organization does not begin with the tool, but with the person who will use it. When we introduce employees to the technology first, ask them about their needs, make them partners in development, and train them in parallel with implementation, they turn from resisting change into leading it.
If your organization is preparing to introduce AI, I would be glad to design with you a preparation and training program suited to your teams and their different levels.
