Notice No. 07-25
The U.S. Department of Labor released Training and Employment Notice No. 07-25 on February 13, 2026, introducing its Artificial Intelligence Literacy Framework as voluntary guidance for workforce and education systems.
https://www.dol.gov/agencies/eta/advisories/ten-07-25
The purpose, plainly stated, is to encourage expanded AI literacy training and to offer a structured resource for program design. It sounds straightforward. And yet, reading it, I felt that odd mix of urgency and optimism that has followed nearly every major technology announcement in the last year.
At its core, the framework defines AI literacy as a foundational set of competencies that enable individuals to use and evaluate AI technologies responsibly, with particular attention to generative AI. Not every worker must become a machine learning engineer. That is not the point. The goal is a baseline understanding. A working fluency. Literacy that lets someone draft an email with AI assistance and still know when the tool is confidently wrong. Because it can be wrong. Convincingly wrong.
The document situates this effort within broader federal initiatives, including Executive Order 14277 and earlier DOL guidance, such as TEGL 03-25, both aimed at expanding AI education for youth and adults. It also aligns with Department of Education actions in July 2025 to advance AI use in education. This is clearly not meant to be a solitary memo drifting into bureaucratic space. It is part of a larger movement, one that has gathered speed over the past year as generative AI tools became almost impossible to ignore.
The framework outlines five foundational content areas. First, understand AI principles. Workers are encouraged to grasp how AI systems rely on pattern recognition and probabilistic outputs, how training differs from inference, and why hallucinations occur. That word alone still feels strange in a policy document. Hallucinations. As if the software had wandered into a dream and returned with half-remembered facts.
Second, exploring AI uses. This means exposure to practical applications, such as drafting documents, analyzing data, generating creative content, or supporting decision-making. AI platforms are efficient, almost dazzling; however, they also can misinterpret key points depending on the context. Efficiency is seductive. Accuracy requires vigilance.
Third, directing AI effectively. Prompting with clarity, providing context, iterating on outputs. The framework emphasizes that coding skills are not required, but thoughtful instruction is. It is less about technical wizardry and more about disciplined communication. Like giving instructions to a quick, very literal intern.
Fourth, evaluating AI outputs. Workers must verify facts, assess completeness, detect logical gaps, and apply human judgment. The framework is firm here. AI should assist, not replace, accountability. Humans remain responsible. Full stop.
Fifth, using AI responsibly. This includes protecting sensitive information, following workplace policies, avoiding misuse, managing risk in high stakes settings, and maintaining accountability for final outputs. The emphasis on responsibility feels especially current, given the recent public debates over data privacy and AI-generated misinformation. It is almost as if the framework is quietly responding to headlines without naming them.
Beyond content, the framework sets out seven delivery principles. It calls for experiential learning, embedding instruction in real workplace contexts, building complementary human skills such as critical thinking and creativity, and addressing prerequisites like digital literacy and broadband access. That last point matters more than many admit. AI literacy presumes internet access, device familiarity, a certain comfort with screens. Not everyone begins from the same starting line.
Other principles include creating pathways for continued learning, preparing enabling roles such as managers and career counselors, and designing programs for agility so content can develop as AI tools change. Agility is the word of the decade, isn’t it. And perhaps rightly so. AI systems evolve in months, not years. Training programs cannot remain static without becoming obsolete.
The framework is explicitly described as a starting point, open to revision based on stakeholder input and technological change.. No one can pretend that a February 2026 document will perfectly capture AI’s trajectory for the next decade.
In summary, the Department of Labor’s AI Literacy Framework establishes a national baseline for AI competency across workers, employers, educators, and agencies. It defines what AI literacy means, identifies essential content areas, and proposes delivery strategies that are practical and adaptable. It is ambitious but grounded. Technical yet accessible. A blueprint, though not a finished building. And perhaps that is fitting. AI itself feels unfinished. Expansive. Occasionally brilliant. Occasionally baffling. The framework does not promise certainty. It promises preparation. Which, in this moment, might be the most realistic offering of all.
In the past eight years I have spent the majority of my time building up a busy divorce and custody practice. We now have six attorneys and conclude hundreds of...
By Howard IkenBruce Meyerson is a Board Member of the AAA-ICDR FoundationThe legal system has historically been civilization’s shared hope for accessing systematic justice. Regrettably, the COVID-19 pandemic has made judicial activity...
By Bruce E. MeyersonFrom the Mediation Matters Blog of Steve Mehta.Dale Carnegie once said “Flattery is telling the other person precisely what he thinks about himself.” This statement is probably very telling about...
By Steve Mehta
