DOL Releases Groundbreaking AI Literacy Framework: 5 Early Best Practices

Most organizations have come to terms with this: Artificial Intelligence (AI) has or will change how their employees work.

Now there are guardrails to help get that work done. The U.S. Department of Labor (DOL) has released the Artificial Intelligence Literacy Framework. With that, the DOL has essentially declared that AI literacy is a foundational workforce competency across industries.

We’ll let the framework explain itself: AI is rapidly reshaping the economy and transforming how work gets done. From offices to manufacturing floors, hospital wings, classrooms and more, AI tools are being adopted across sectors, changing how tasks are completed and how decisions are made. In an increasingly AI-driven economy, every worker will need baseline AI literacy skills to succeed, regardless of industry or occupation.

Origins of AI Literacy Framework

The AI Framework stems back to President Trump’s Executive Orders aimed at technology, and specifically in response to Advancing Artificial Intelligence Education for American Youth.

It could be welcomed guidance by employers who are overwhelmed by AI from every angle. The framework outlines five skill areas employers will want employees to master: 

• Understanding AI principles 
• Applying AI in real-world contexts
• Directing AI tools effectively through promotion and iteration 
• Critically evaluating AI outputs, and
• Using AI responsibly and ethically.

“The framework recognizes that AI literacy is becoming a baseline workforce skill and no longer just one that’s ‘nice to have’,” says Mark Quinn, Senior Director of AI Operations at Pearl, a super-agent powering the independent professional economy. “Engineers and developers aren’t the only ones who need AI skills. Workers across all professional industries, and especially within high-stakes industries, like law, finance, and healthcare, need to understand not only how to use AI, but how to use it effectively, safely, and compliantly.”

That’s partly why the guidance also stresses that how employees are trained is as important as the training itself. Training programs should be:

• Experiential and contextualized to industry needs
• Agile in design, and
• Aligned with labor market demands.

Train Employees for AI Literacy

Ahead of the guidance, SHRM had the research that showed AI literacy training worked. In organizations where AI was adopted by the end of last year, only 7% of HR professionals reported layoffs due to AI. Meanwhile, 24% said they created new roles, 39% said they shifted worker responsibilities, and 57% reported new upskilling or reskilling because of AI. 

The good news: The DOL issued a framework. These didn’t issue mandates. Your organization now has guidance to help employees adopt and master AI responsibly and effectively.

Here are five early best practices:

1. Create and Communicate Balance

The guidance helps organizations create governance around usage, security and safety policies from when AI is first introduced.

“There must be clear communication and a balance of ensuring safety without being overly restrictive, because employees may use AI secretively out of concern that its usage is wrong,” says Quinn.

2. Integrate Rather Than Initiate

“Build AI literacy into how people are already working, rather than treating it as a separate initiative that feels mandatory and generates unnecessary fear,” says Quinn.

In fact, some tech companies embed AI literacy into hiring, where candidates are considered based on their curiosity and adaptability for adopting the technology. At other big tech companies, AI literacy has been embedded into performance reviews, where employees are rewarded not for whether they are using AI, but for how they are using it.

3. Focus on Accuracy

Going back to two of the five key points in the AI literacy framework — 1) critically evaluating AI outputs, and 2) using AI responsibly and ethically — organizations will want to focus on accuracy early and often.

At Pearl, they already hold mandatory AI training for accuracy.

“These measures ensure that the fundamental understanding of AI’s shortcomings and the need for verification from a human expert is baked into every employee’s daily approach to their work,” says Quinn.

4. Find Champions, Share Results

AI learning and development can be exciting and intimidating. Again, with AI literacy guidance, organizations have the room to teach and build in ways that work best for their cultures.

In many organizations, AI isn’t just rooted in the IT development. When champions come from different areas, employees are more likely to embrace AI and its possibilities.

For instance, at Pearl, they provide space for a shared learning experience with an “AI Champions Program.”

“Department leads and senior leaders commit a portion of their time to learning, experimentation, and delivery. This delivers clear business value through real projects that streamline, automate, and improve work with AI,” Quinn explains.

5. Incentive AI Literacy

People fear AI will take their jobs. So they might run or hide in fear. But they’ll probably jump at the chance to embrace AI if they’re incentivized!

Even better, Quinn says, “Management can offer these payouts to employees at all levels and can pull funding for them from the AI-driven cost savings employees discover and implement.”

But really, it doesn’t just have to be about money. You might want to focus on non-monetary incentives such as career growth, improved workflows, public successes or leadership roles on AI projects.

“A McKinsey study shows that the most effective driver of AI adoption for employees is formal training and skill development,” Quinn points out. “Connecting AI skills to tangible rewards like promotions, internal certifications, or access to specialized learning opportunities shows that mastering AI has professional value, encouraging workers to thoughtfully engage with AI and strive for success.”

Case Study: Pearl’s ‘How I AI’ Training Sessions

Pearl created an early version of AI Literacy training. We’ll let Mark Quinn explain so you can pull from their best practices.

At Pearl, we host ‘How I AI’ training sessions for our employees where we share practical, everyday use cases that they can apply to their own work, from writing and collaboration to building agents and workflows.  

One of the sessions I host uses a ‘Hallucinator’ agent I built to demonstrate how AI hallucinates. It’s an interactive session where our employees can have conversations with the agent and, from there, identify its inaccurate outputs. There is even a ‘teaching mode’ built into the agent, which helps users understand hallucinations by providing clear explanations and examples. 

We’ve also held sessions on how to build your own personal AI tutor. Participants learn how to not only create their own personal learning assistant but also how to make learning fun for anyone. The goal is to help participants understand how they can use AI to customize materials, create extended supporting materials, and explain complex industry concepts.

Overall, the goal with these ‘How I AI’ sessions is to make learning approachable.

One of the first sessions I ever hosted was on cooking with AI, where participants learned how to create a custom agent that can serve as their personal sous chef. While these explorations may seem outside any professional use case for AI, they help participants internalize key prompting and AI problem-solving skills. By using AI to create recipes and accomplish other everyday life tasks, participants also learn how to apply it to their own work. The cooking exercises made a potentially intimidating lesson for some more approachable because they introduced how to use AI and its possibilities for their work through low-stakes scenarios. 

Overall, adopting an AI literacy framework will come with its share of failures before it’s successful. You have to keep in mind, that AI, much like people, has its strengths and weaknesses, so the goal should not be to perfect AI. Rather, it’s about developing the understanding required to leverage AI’s value drivers while managing its limitations. To design the best-suited framework, company leaders must stay open-minded and experiment, test, and iterate until they find what works best.