“Eneke the bird says that since men have learned to shoot without missing, he has learned to fly without perching,” writes Chinua Achebe in his 1958 novel Things Fall Apart. The line resurfaced at a conference, where educators debated how artificial intelligence reshapes learning.
Adapting education for an AI era
Ghana’s Country International School teacher Nana Amakah Otu opened the session by quoting the bird. “Like Eneke, education has to adapt,” she said, noting that AI can generate essays, summarize texts and translate languages, forcing schools to rethink core competencies.
Conference speakers agreed that the question is no longer whether students will encounter AI, but how they will use it without surrendering critical thinking. Sessions emphasized guiding learners to treat AI as a tool, not a substitute for intellectual effort.
Balancing performance and understanding
Professor Rose Luckin, CEO of Educate Research Ventures, presented data suggesting that heavy reliance on generative AI reduces brain activity during essay writing and weakens recall even after the tool is withdrawn. She warned that “the risk is not AI itself. The risk is unmanaged AI.”
Luckin’s findings echo neuroscience research that the act of grappling with a problem, not the polished product, builds lasting skills. When students click a button for a finished answer, the mental pathways that support deep learning remain underused.
Rather than banning AI, experts urged schools to embed it within curricula that reward the learning journey. One teacher at the conference described the need to celebrate the “messiness of learning,” highlighting that struggle often precedes insight.
Key human skills identified as resistant to automation include metacognition—awareness of one’s own learning processes—meta-knowing, and perceived self-efficacy. These abilities enable students to evaluate AI output and decide when to rely on it.
Research presented by Sarah Hughes and Silvia Pirola of Cambridge showed that while teachers and learners recognize metacognition as essential, they also view it as the hardest skill to teach. Their study linked higher-order thinking requirements in Cambridge International AS & A Levels to this very challenge.
In a breakout, Andrew Field asked educators which future skills grow in importance when AI supplies instant answers. The consensus returned to metacognition, with participants calling for formal assessments that measure students’ self-regulation and reflective capacities.
One practical approach emerging from the discussions is “AI reflection mapping.” Teachers ask students to generate an AI response, then dissect it, noting strengths, weaknesses and any misconceptions, and to back their critiques with evidence from textbooks or reliable sources.
The qualifications require simultaneous application of deep knowledge and metacognitive awareness, prompting learners to monitor their understanding as they work through complex scenarios.
As the conference concluded, AI may be able to generate answers instantly but understanding how we learn remains uniquely human. In this rapidly changing world, students need to be able to understand and use AI efficiently and effectively in order to leave school with a skillset that prepares them for university and beyond, and it is up to educators to help them develop these skills.
Leave a Reply