Generative AI Adoption is Dwindling – How Do We Fix This?

ChatGPT was all the rage in early 2023, but now, it’s almost become a punchline in some productivity circles. First launched in November 2022, the tool reached 100 million users in just two months. But if new research is to be believed, those users haven’t stuck around – at least, not for routine use. Only 2% of U.K. consumers and 7% of U.S consumers say that they’re regularly using generative AI tools like ChatGPT. That’s despite almost one-third (30%) of the public being aware of such technology. ChatGPT has an adoption problem that would keep most technology leaders up at night. So, how can this be fixed?

From Hot to Hype

When generative AI tools hit the mainstream, we saw users come up with all kinds of prompts aiming to improve their work productivity, research speed and even their diets and fitness. In its golden era, you wouldn’t have been amiss predicting that we’d use AI in every aspect of our lives, from morning to night. Yet, this hasn’t proven true, and a huge reason behind this is a lack of application-specific skills among users. The plethora of prompt-engineering cheat sheets, still being published years after launch, already show that people are struggling to get their heads around how to correctly apply ChatGPT skills to their everyday work.

Skills are the Enabler

Whether you are an IT leader looking to upskill your employees to use the latest crop of AI solutions or a vendor seeking ways to improve customer enablement, sales effectiveness and product adoption, skills are the foundation to everything. No skills equals no confidence – in the case of generative AI, a lack of understanding how to apply such tools to your work for maximum productivity is seriously undermining adoption, causing distrust in outputs and hindering results. Get the skills right for your chosen technology and the rest is a lot easier to put into place.

Where to Start

Begin by understanding your starting point. Get a baseline reading for your chosen group’s skill level (customers, partners and resellers, a whole workforce or specific teams). For employees, human resources (HR) and learning systems will likely have some helpful information and then you can assess skill level through real-world scenario testing and project and peer feedback. For customers and partners, you may have to rely on in-the-moment performance-based testing. Some organizations will rely on multiple-choice quizzes to assess skills, but these can be easily gamed by looking at multiple screens for the right answers, or guesswork.

Get Specific with Training

Once you have a good idea of what relevant technical skills your group has and where they need to improve, you can set about upskilling them for your technology implementation.  

Not all training is equal, however. Content- and knowledge-based learning will get people to a basic level, but for true job readiness and confidence in the heat of the moment, you need to put skills into practice. For technology, that is now relatively simple. Gone are the days of hard-to-maintain sandboxes and laggy virtual machines. Modern virtual lab environments can be created and deployed within weeks, continuously maintained and available as soon as a learner needs access to them. Giving people a safe space to practice using your chosen technology, in the specific ways that your business needs them to understand, is the only way you can ensure they are ready to apply their skills to the job.

You can even add in a validation element by setting specific tasks or creating scenarios like “.”

Skill Levels Vary

With the rapid pace of AI development, you need your team to continuously upskill so their abilities keep up with business needs and changing demands. A comprehensive understanding of AI and its applications within your business can also support responsible use and governance. Everyone needs a certain level of AI literacy to effectively work alongside, govern, oversee and implement new innovations (and to know which AI tools aren’t relevant for their needs).

That segues nicely into the next point: Not everyone will need to be an AI master. As McKinsey and Co. calls it, your employees and customers will be divided into “takers, shapers and makers” depending on their role and how they use generative AI. Self-explanatory, but takers, as the primary user of an AI tool, could get away with foundational AI training. The more nuanced and influential shapers and makers will need true skills mastery. This can only be achieved through deep, targeted training with a real-world, experiential element. Dividing groups into required skill levels (basic through master) will help to tailor learning and optimize your training resources.

The final puzzle piece is your culture and workplace attitude toward training. AI upskilling today isn’t a one-and-done effort but something that needs to happen consistently with ongoing improvement over time. When learning is a habit, people are more readily able to absorb new information and know where to go to build a new skill when the next ChatGPT launches. Adoption won’t hit a stalling point, at least from a skill-enablement standpoint, if people have relevant job-ready skills from launch day onward.

Skills Come First

Skills unlock all of the other factors that drive technology adoption from delivering and proving value to getting senior stakeholders on board. People need their skills in place before they can even consider how to apply a new solution to their role and workplace for the best results.