Why AI Upskilling is More About Change Management Than Technology

More than half of organizations and businesses in the Nordic countries are already utilizing AI, with two-thirds planning to expand their use of AI over the next three years. While this growth path might not surprise many, it becomes more interesting when compared with a key challenge: 28% of organizations cite a lack of digital skills as a major barrier to fully leveraging AI, and 35% currently lack digital skills. This highlights the critical importance of upskilling employees to achieve organizational AI objectives. However, this task is less about the technology itself and more about securing leadership buy-in and changing organizational habits. 

In the past (and that’s literally just 3-5 years ago), IT projects were mainly handled by the IT department, which decided how to implement new platforms, tools, software, or hardware across the organization. This approach relied heavily on hope for successful outcomes. Those days are gone. Today, IT and new technologies require a broad range of organizational capabilities for effective deployment and usage.  

Like other strategic projects, such as sustainability, they have moved from the corner of the organization to become a cross-departmental interest, anchored at the top management level. 

Empowering AI transformation through top-level anchoring  

If a project is of strategic interest (and most AI projects, are) it needs anchoring at the top management level. Without a clear C-level mandate, AI will not be successful. The mandate should include a strategy to close the skills gap I just mentioned. And thus, management needs to ask the crucial questions: What are our AI capabilities today? Where do we want to be in x years? How can we close the gap?  

This analysis will help organizations strategically plan their upskilling efforts. A report from Digital Dogme about AI-readiness states that while 42% of organizations make AI-related decisions at the management level, 75% lack a strategy for using AI. This likely indicates that many organizations also lack a strategy for AI upskilling. 

Set AI free! 

Once a clear mandate from the top management is established that defines the AI playing field (governance model, resource allocation, compliance, KPIs, etc.), setting AI free becomes the next step. This may sound frightening, but it’s essential. Management can only control AI to the extent that everyone understands its importance and potential benefits for both the business and employees.  

But that will not do the trick alone. Changes must be driven by passionate employees eager to explore what AI can do for them and their jobs. These employees should be the first to receive new tools or licenses, like Copilot for M365 or whatever tool you are testing – they are your “employee zero.” If you start by providing new tools or technology to people who have no interest in them, they will never adopt them. If someone is not using their license, take it back and give it to someone who will.  

Identify immediate reward projects 

Next, identify AI use-cases that offer immediate rewards, such as improved workflows or new ways to summarize meetings or actions. These quick wins provide users with a sense of low investment and high reward, encouraging change. Why? Because people are people. We are simply wired to make things as easy for ourselves as possible. If we see that a given action gives us a reward that outweighs the mental cost of changing habits, we will do it. Once you have a group of ambassadors and high-reward AI use-cases, roll them out across the organization. Encourage employees to share their experiences and results. 

Involve your communications department to maintain a high flow of information. The marketing concept of the “rule of 7” suggests that people need multiple exposures to a product or service before deciding to purchase it. My hypothesis is that the same applies to new technology. Over-communicate, nudge, and gamify your efforts. 

But don’t we need official training and AI courses? 

Yes, you do. But both training and rollout should happen simultaneously. Introduce and test new AI tools, get successful use-cases flowing, and then provide function or role-based training. This approach works better when there is a recognizable frame of reference – like a new tool employees have already tried and are now expanding upon. If you’re interested in exploring learning opportunities to upskill your workforce, you can find our dedicated site with all the training tools you need to close the gap below. 

The cost of inaction 

Upskilling your entire organization is undoubtedly challenging. It requires time and sometimes the acceptance of certain compromises. In the past, “shadow IT” referred to the use of new tools by employees without official guidance, often leading to risks and non-compliance. IT departments typically disliked shadow IT due to its hidden risks. However, when it comes to AI training and upskilling, you might need to rethink this perspective.  

If top management establishes a clear framework, shadow IT can transform into “innovation IT.” Many organizations are still figuring out how to fully harness AI. Exploring AI’s potential within a structured environment can boost success and uncover valuable new use-cases for the entire organization. This approach requires both courage and patience, but the cost of inaction is greater than the benefits gained from fostering a curious and AI-skilled organization. 

Want to learn more? 

• Go to our AI upskilling website here: http://www.microsoft.com/learn

• Read the report “Are Nordic organizations ready for AI” here: https://digitaldogme.dk/viden/are-nordic-organisations-ready-for-ai  

• Learn how DSB prepared its organization for Copilot for Microsoft 365 here: https://pulse.microsoft.com/da-dk/transform-da-dk/manufacturing-da-dk/fa1-dsb-copilot-for-m365-kan-skabe-unik-vaerdi-for-vores-medarbejdere-men-er-endnu-ikke-en-schweizerkniv/