The New Imperative: “Learning How to Learn”
Google DeepMind’s CEO Demis Hassabis has argued that “learning how to learn” will be the most important skill for the next generation. Speaking at an event in Athens, he explained that the pace of AI-driven change makes continual adaptability essential. “The only thing you can say for certain is that huge change is coming,” Hassabis noted, highlighting the need for meta-skills: understanding how to learn and how to approach new problems, rather than relying only on static knowledge.
This perspective resonates widely. Educators and technology leaders alike stress that “learning to learn” is becoming the most in-demand skill, because success increasingly depends not on what you already know, but on how quickly you can acquire and apply new knowledge. In short, the ability to adapt has become as important as knowledge itself.
Information at Our Fingertips – and the “Google Effect”
At the same time, we live in an era of ubiquitous information. Almost any fact can be summoned instantly through a search engine or an AI assistant. This convenience leads us to rely on digital tools as extensions of memory—a phenomenon known as the “Google effect” or “digital amnesia.” Psychologists have found that people are less likely to remember information when they know it will be available later. In other words, we increasingly remember how to find information rather than the information itself (Sparrow et al., Science, 2011).
This concern is ancient. Over two millennia ago, Socrates warned that writing would “introduce forgetfulness into the soul,” because learners would no longer exercise memory, relying instead on written words. Writing, he said, was “a potion for reminding, not for remembering.” Today, similar criticisms are leveled at search engines and AI: they give the appearance of wisdom without requiring actual understanding (Plato, Phaedrus).
The Paradox of Memory in the Digital Age
Here lies a paradox. If knowledge is always available, why bother remembering? Some argue this is liberation: it allows us to focus on creativity and problem-solving instead of rote memorization. As one technology commentator put it, “Why memorize dates or formulas when those will always be right there with us, available for instant recall?” (Mike Elgan, eWeek, 2011).
But there is a cost. Psychologists warn that heavy reliance on AI and digital tools can erode our natural memory skills. We practice what is called cognitive offloading—outsourcing mental tasks like navigation, calculation, or recall to devices. Over time, this weakens the very mental “muscles” we neglect (Barbara Oakley et al., The Memory Paradox, 2025). Memory, however, is not just storage: it is the active process of retrieval and practice that builds deeper understanding. Without a base of remembered knowledge, it becomes harder to think critically, make connections, or spot errors.
Crucially, memory and learning-to-learn are not opposites. They are complementary. To learn a language quickly, one must remember vocabulary; to master advanced mathematics, one must recall basic formulas. Internalized knowledge provides the scaffolding on which adaptive thinking can rest. As education researcher Carl Hendrick has observed, “AI can simulate intelligence, but it cannot think for you. That task remains, stubbornly and magnificently, human.”
Implications for Education
In the past decades, schools have shifted away from rote learning toward problem-solving skills, often allowing calculators or open-internet learning. Now, with AI in the mix, researchers argue that completely abandoning memorization was a mistake (Paul Bennett, Policy Options, 2025). Oakley and colleagues point to two forces behind declining cognitive ability: first, an educational retreat from practice and memory drills; second, growing dependence on technology for basic mental tasks.
Educators are now urged to pursue “cognitive complementarity”—balancing strong internal knowledge with thoughtful use of external tools. Practical strategies include:
- Retrieval practice and spaced repetition: Strengthening recall without digital aids ensures long-term retention.
- Delayed tech reliance: Encouraging learners to attempt recall or reasoning before consulting AI.
- AI-aware curricula: Using AI as a complement—editing, extending, or fact-checking—rather than as a substitute for original thought.
The aim is not to turn back the clock, but to produce thinkers who can wield tools like AI effectively because they also command knowledge of their own.
Broader Societal Implications
The stakes go beyond classrooms. A society that forgets how to remember risks losing vital skills: mental arithmetic, wayfinding, or even common knowledge. This can make individuals helpless without devices—and overly dependent on platforms that deliver information. If only a handful of AI systems supply answers, they become gatekeepers of truth. Without independent memory, our ability to question and verify diminishes.
This raises profound questions: What does it mean to be wise in an age where information is always at hand? Does the ability to appear knowledgeable—by instantly consulting an external source—count as knowledge at all? The answer may lie in balance: embracing the marvels of AI and instant knowledge, while also recognizing that memory remains the bedrock of understanding.
Conclusion
Demis Hassabis is right: learning to learn is the defining skill for the future. But there is a danger if we neglect the role of memory. In an age of endless information, remembering becomes paradoxically both harder and more important. For without a store of knowledge in our minds, our ability to think, to question, and to truly learn may fade. The future belongs to those who not only adapt quickly, but also preserve the inner strength of memory—the most human form of intelligence we have.