Artificial intelligence (AI) is permeating every part of our lives, from smart home devices and phone assistants to workplace co-pilots and agentic AI. Its impact on the workforce is unparalleled, but still increasing, much like how machinery in the first and second Industrial Revolutions reshaped the way we worked, lived and socialized. Today’s workers are navigating an ever-changing future and, worryingly, research is showing that their skills aren’t keeping up.
Supply, Demand and Perception
Despite the rapid adoption of AI, there is a troubling gap between how many professionals believe they can work with AI and those who possess the proven, necessary skills for an AI transformation. Given that we don’t fully understand what AI-enabled solutions may launch over the coming years, all workers need to continuously upskill and reskill as AI technologies evolve.
Skill development is only one side of the AI skills equation. How organizations assess and validate the workforce is equally important to understanding the capabilities of a person, team, or channel partner. Are the people upon whom your success depends ready to apply AI to tasks and projects? This challenge clearly extends to hiring too, as 32% of individuals admit exaggerating their AI expertise on their resumes and 30% during interviews.
Put another way, the current methods for acquiring (e.g., build and hire) and validating skills aren’t enough to give full confidence in your organization’s AI capabilities.
Rethinking Training in the Age of AI
AI not only transforms the tasks we perform but also how we approach learning itself. By automating the recall of facts and basic concepts, AI allows workers to allocate their mental energy to higher order thinking such as organizing, analyzing and creating. Content and knowledge-based training can cover lower-order tasks like recalling and recognizing facts, but for true strategic decision-making and ideation you need a performance-based element that allows people to practice, fail, experiment and explore without fear (of damaging operations, data or technologies).
Experiential learning has been around since the early eighties, but it’s about to hit its renaissance. Real-world experience is the only way you can train people to use AI and work in an AI-powered reality. They can use new tools, processes and methods to accomplish the outcomes your organization needs. That can be done on-the-job through stretch assignments, secondments, temporary deployments and apprenticeships, or virtually, using labs. Hybrid options exist too, combining mentoring and virtual labs, for example, or integrating scenario-based environments into learning pathways.
Trusting Your People’s AI Skills
The other side of the equation, skill validation, then comes into play. You need to be confident that your existing workforce, new hires and extended enterprise of partners and customers, can work with AI within the context of specific workstreams. That will be different for every project, department, and organization. So, if you hire a data scientist from one academic background or industry, you still need to assess if their skills are transferable to your exact needs. Likewise, you’ll need to test your workers within set boundaries and with specific features such as your data governance and AI compliance rules, your technologies, and your data.
Skill validation tells you how effective your training has been in equipping people to use AI in the real world (along with other technologies and skills). Tailoring it to your organization’s processes, policies, technologies and other operations is key to ensuring people can perform a skill as needed, for your specific use case.
Symbiotically, AI itself can help with validating skills either through assisting learning teams and lab developers to write the assessment criteria (timings, tasks that need to be completed, tools to be used, rules to be followed and so on) for your validation exercises, based on a job or project description. Or AI can support by visually confirming that a task has been completed to the necessary standard. For instance, if the assessment wants someone to clean a set of data ready for algorithm training, the AI feature can validate specific tasks have been done such as locating and using the right data set, putting the data into the correct format and saving it in the appropriate place with the necessary user permissions.
Best Practices for Training and Validating AI Skills
To begin shifting the way you train and assess people’s skills in the wake of AI, here are five best practices to focus on:
1. Make it relevant.
Take the time to clearly understand what people need to do on the job. From this, tailor training to specific tasks and technologies that each individual will use. This ensures participants see immediate relevance and applicability.
2. Make it easy to access.
Embed training into day-to-day activities rather than isolating it as an “off-work” task. If individuals are learning, then doing, throughout the course of their workday they instantly apply their newly learned skills, adopting the methods into their routines and increasing the likelihood of skills being remembered.
Likewise, consider how different groups may access or be unable to access your training. If most of your experiential learning is in-person based, those unable to travel due to budget, geography or caring commitments are disadvantaged. Use a variety of methods, offline and online, to increase the scenario-based aspects of your training.
3. Make it interactive.
Studies show that the human brain can focus on training from five to fifteen minutes before losing focus and needing time to consolidate the knowledge learned. This can be extended slightly with interactive elements that help with this consolidation and application phase. Therefore, plan for your training to have 15-minute task-focused blocks with interactive elements to enhance engagement and retention, with your overall session lasting anywhere from 15 to 45 minutes.
4. Make it outcome based.
Focus your skill validation on the outcomes achieved by a learner and not the facts or basic concepts that they can relay. For example, assess that they can use a generative AI tool to effectively draft a follow-up email to a sales prospect or that a developer can use a co-pilot to troubleshoot and identify incorrect code.
5. Make it shareable.
Connect your experiential learning and skill validation tools and processes to your wider learning strategy and business intelligence tools. The skill validation data, in particular, will be a rich source of accurate, timely and high-trust data for skills-based decision making in the future.
Reshape Your Future — or AI Will
AI continues to march on, reshaping the professional landscape and the skills required for it. Adapting your approach to training and skill validation is imperative, not just today but continuously, as new AI advances come to market. Develop a culture of continuous skilling, validation and understanding your skill supply and demand, and your organization will be well-placed for every co-pilot, AI agent, Alexa, Gemini and Hal-9000 that comes to the workplace.
