The impact of AI on education: broader, more sensible curricula

This post originally appeared on the Christensen Institute blog and is republished here with permission.

Key points:

The world around us is changing profoundly and education must adapt to it. In a time of increasing uncertainty, a smart strategy would be to protect against unrest by adopting versatility.

Today, a K-12 education must equip students with the skills to confront life's challenges, ranging from social and political issues (global warming, pollution, inequalities, etc.) to the evolution of technology ( social networks and now particularly artificial intelligence). Such as, educating is NOT training (with a respectful nod to former MIT professor Woodie Flowers, RIP); Education is broad and related to life in general.while training is limited and job-related (and partially begins in high school). Of course, education and training are necessary at some point, but they should not be confused, as they have different objectives (psychosocial-focused for education, and economic-focused for training).

Since the future is unknowable, cultivating versatility is a wise and appropriate strategy; Consider it “protection against all eventualities.” Using the Swiss Army knife analogy, it is best to equip students with a broad set of tools that can be sharpened as circumstances require: poet, doctor, painter, and physical.

How wise? Redesigning the 'What', not just the 'How'.

Wisdom is, more than ever, the goal of education. But to get there, both standards/curriculums (the “What”) and pedagogy (the “How,” which is not covered in this short blog) need to be redesigned, as there are significant gaps between emerging needs and practices. current.

The fact that:

Education has not yet fully adapted to the information age; for example, although it is called “STEM”, only “St_M” is taught in K-12: very little technology and no engineering. Twenty-five years after the beginning of the Internet era, which David Houle amazon.com/Entering-Shift-Age-Information-Transformation/dp/1402272170″ target=”_blank” rel=”noreferrer noopener”>called the era of changeModernity requires rapid adaptation to changing information and effectively addressing a diversity of languages, cultures and lifestyles. And with the era of ai, the accumulated deficit of the last two eras (industrial and computer science, neglecting deep curricular adaptations) returns with even greater force, increased by a new set of challenges.

As a result of this inertia, some will argue, perhaps to moderate their cognitive dissonance, that the “what” doesn't matter “as long as it is learned.” At the Center for Curriculum Redesign (CCR), the organization I direct, we deeply disagree: why focus teaching on old content if there are better options available? For example, why waste time learning advanced trigonometric functions that are important to very few and have been largely automated, instead of data science, which is useful in many disciplines and is in high demand?

The four dimensions matter:

Chris Dede of Harvard resume gets the situation right: “Current curriculum and high-stakes testing often prioritize fostering skills that ai excels at, such as calculation skills that involve calculative prediction and formulaic decision making. However, ai cannot easily replicate human judgment, which is a deliberative, flexible and contextual thought process based on experiential knowledge, ethics, values, relationships and culture.”

As described in the 2015 CCR book. Education in four dimensionsThis means paying attention to everyone. four dimensions of Education: Knowledge, skills, character and meta-learning.

Knowledge: Declarative knowledge is challenged more than ever by large language models (LLMs), which amplify historical trends (amazon.com/Plato-English-translation-Part/dp/1175322563″ target=”_blank” rel=”noreferrer noopener”>scriptsbooks, Internet, search engines). As explained above, that doesn't mean humans don't need basic knowledge, it means they need to be much more discriminant about what is essential and relevant.

Furthermore, and counterintuitively, there is a need for a broader set of declarative knowledge to respond to the need for versatility. David Epstein, author of amazon.com/Range-Generalists-Triumph-Specialized-World/dp/0735214484″ target=”_blank” rel=”noreferrer noopener”>Range

contrasts “kind” and “wicked” learning environments, and explains that while structured and predictable “kind” environments may encourage early specialization, real-world “wicked” environments often reward a wide range of experiences. It also emphasizes that a sampling period, in which one explores several interests before settling on one, can lead to greater career satisfaction and success. Epstein opposes early specialization in education, stating that it can limit children's ability to explore their potential and adapt to new situations; and investigates “quality of match”: the match between one's interests, skills, and career, and how a wide range of experiences can improve it.

However, it's crucial to establish an 'and' mindset at this stage: according to IBM Research T-shaped modelIt is perfectly possible to develop depth AND breadth, not one or the other: experience AND transfer. This is the CCR's position: LLMs increase the pressure to teach more conceptual knowledge (basics) and procedural knowledge (projects). Success in medical, legal and other tests highlights the extent to which these tests rely on the memorization of declarative knowledge (to be fair, along with some deductive abilities).

Competencies:

• Skills: Challenged and enhanced by ai.
• Character: Some are still significantly human (e.g. ethics) and need to be supported, while others are helped. and pushed (e.g., curiosity).
• Meta-learning: Learning how to learn is more fundamental than ever, as are metacognition and meta-emotion.

In addition to the goals of modern education outlined in the Venn diagram above, there is also a growing need to personalize education. This personalization comprises four drivers: motivation, identity, agency and purpose, of which motivation and purpose will remain essentially human.

In short, to fully realize the transformative potential of ai in education, we must rethink what we teach and how we teach, ensuring that students are not only prepared for the jobs of tomorrow but also equipped to navigate life's complexities with wisdom. Adaptability and continuous, self-directed learning are key. This also means fostering depth and breadth of learning, where students develop specialized knowledge and activate transfer, while acquiring the skills, character, and meta-learning capabilities necessary to thrive in an unpredictable world. By prioritizing these four dimensions (knowledge, skills, character and meta-learning) and by personalizing education and improving student motivation through the three drivers of identity, agency and purpose, we can create an educational system that is truly responsive to the demands . of the 21st century and beyond.

The innovative potential (by Thomas Arnett, senior researcher at the Clayton Christensen Institute):

Charles Fadel's commentary is a valuable and much-needed contribution to the broader conversation about the impact of ai on K-12 education. While many discussions on this topic remain high-level, Charles and his colleagues have taken the critical step of providing a comprehensive and detailed analysis of what specifically needs to change in terms of the curriculum and student learning experiences. This blog post offers just a glimpse of his insightful ideas, and I highly recommend that readers who are intrigued by these concepts explore them further by obtaining a copy of his book. ai/” target=”_blank” rel=”noreferrer noopener”>Education for the era of AYO.

However, when contemplating the significant changes that Charles advocates, an important question arises: What will it take to get educational systems to adopt these changes in curriculum and pedagogy? On the one hand, as the challenges posed by ai become more pressing and change accelerates, it may be possible to muster the political will to build a society-wide consensus that drives legislative action at the state and federal levels. But waiting for that political push, which may or may not align with the timeline we need, should not be our only strategy.

In the short term, I believe these changes may find their greatest momentum among those who are actively working to build new models of schooling within their countries. new value networks. Rather than waiting for conventional education systems to undergo a massive shift in priorities, the initial footholds of change can emerge more quickly in spaces that incubate unconventional models of schooling, such as hybrid virtual learning, microschools, and alternative education. . These environments often have the best opportunities for escape from limitations of conventional education and integrate forward-thinking ideas like those Charles describes. As these new models gain momentum, they can serve as testing points that can eventually influence broader systemic change.

This blog is an excerpt modified for the Christensen Institute from the recent book by the Center for Curriculum Redesign (CCR): ai/” target=”_blank” rel=”noreferrer noopener”>Education for the age of ai. For more information, ai/” target=”_blank” rel=”noreferrer noopener”>Click here.