What AI Can't Do: Why The Human Story Still Matters in Education

We believe deep down that the most important skills we can help a student develop are the ones no algorithm can replace. We feel it is crucial that students be able to look thoughtfully at information, not just objectively but also critically. Lacking this kind of analytical thinking, students (and adults) can easily be swayed to believe whatever an external source is telling them is truth. This way of consuming information relegates the individual to 'lemming' status. It dehumanizes and congeals individuality. Living out one's life as another brick in the wall is not a particularly worthy goal.

We are all now well aware of what artificial intelligence is capable of. It's not 'coming', it's here. Whatever we want to think about it, pro and/or con, the smart path forward is to look at education with a keen eye on understanding that what we teach and how we teach it can have tremendous influence on keeping our young people relevant, employable, and valuable in the participatory life of their communities. The big question becomes: what kinds of learning will truly prepare a young person for the world ahead?

One of our favorite organizations is SIMA Academy, a collective which supports and rewards independent documentary film-making around the world. Robert has served as a short-entry judge for SIMA in the past (and before obtaining his M.Ed. had received a B.A. in Film Studies from the University of California-Santa Barbara and worked for a short time as a screenwriter in Los Angeles). One of our feeds recently yielded a piece written by SIMA CEO Daniela Kon Lieberberg. Ms. Lieberberg called it like it is: there is a widening gap between what students can produce and what meaning they actually take away from the process of producing.  To the great detriment of the students, most schools serve as functionaries of the 'produce and move on' style of unbalanced education. The Latin root of the word 'education' holds a duality which has been forgotten in modern, factory-style learning. 'Educare' is the act of training and molding (let's call this the basic fundamentals). 'Educere' on the other hand means to 'lead out', to 'illuminate' (let's call this the meaning behind what we learn). We argue that 'educere' has been largely forgotten since we quashed truly classical education for public mass production. There are some in this world who don't actually want people to think - they just want them to obey. TPA maintains an equal balance of 'educare' and 'educere', and has worked with meticulous care to weave both into a durable and beautiful curricular fabric. The simple point to be made is that AI can make many tasks virtually effortless. Without effort, where's the learning? Where's the individuality? Where's the pathway to meaning? We have to work to prevent AI tools from washing away what it means to be uniquely and wonderfully human, and being human requires effort. This is not to say that AI should be avoided at all costs. It has its place. In full transparency, we use Claude Sonnet to give quick 'blog-ready' shape to our OWN concepts and ideas so that we can quickly and conveniently deliver them to the public. We don't use Claude to create our ideas. We use Claude to digest our ideas, keep to our detailed prompts, and then package that information quickly. It is a tool, not unlike a spelling or grammar checker. As we see it, the slippery slope begins when we, as human beings, turn over the keys to AI and ask it to create something that we, ourselves, should have given full shape to before using the 'tool'. These are also our hard barriers against student misuse. AI can be great as a processing tool - in the same way Gutenberg's printing press began to process the dissemination of the written word in the 15th century - to give shape to a student's original thoughts, but it must not be used for creating those thoughts. If AI can be seen as a potter's wheel, that's all it should be - the tool that helps to keep the process moving so the student can create. At TPA, we require that it be the student's own hands that shape the clay, bake it, paint it, glaze it, and present it to the world. Their hands must learn to shape with personal artistry, judgment, interpretation, and critical reasoning. To do so, however, requires that the delivery of learning constantly reinforces the attainment of these attributes.

The skills that matter most are the hardest to automate

Back to Ms. Lieberberg's piece. SIMA offers a valuable framework that lays out the core competencies that all discerning human beings must develop and keep handy in their tool kit: critical reasoning and ethical judgment. These are heady terms and could easily eat up a full afternoon of teacher training. When we watch our 9-year-old overfilling their cereal bowl for the nth time, it's hard to see the connection sometimes. Thus, it is with great care that these attributes must be developed in our young people. It is our opinion, however, that these concepts are very difficult for teachers to tack on to a traditional program in which they have been trained, and where they are already working on overload to keep up with that school's version of 'educare'. At TPA, these concepts have been baked in from the outset. They are not add-ons; they have been part of the TPA architecture from day one. In other words, an educator who has spent ten years at Ye Olde Elementary-Middle School would likely struggle to adapt to our model. In 'backwards design', a term that gets bandied about regularly at educator conferences, the lesson goal comes first and then the process to try and reach that goal is laid out in aspirational steps. With great care, over five years of passion-driven dedication, TPA has 'backward designed' our entire learning model from the ground up. This is how we've built TPA's deep 'educare'+'educere' into what we truly believe to be among the very top learning models on the planet.

Still from the Italian neo-realist classic, The Bicycle Thief, directed by Vittorio De Sica. / Creative Commons

Back to our story. Visual storytelling sits at the heart of developing the capacities of self-responsible effort, critical judgment and ethical reasoning. Unlike AI-generated visual content or automated summaries (and aren't we absolutely bombarded with this junk nowadays?), a well-crafted film requires its viewers to engage actively; to examine perspective, consider sources, recognize any bias, and ask - often the most uncomfortable part - whose voice is heard and whose is silent. These are all critical and creative processes that a good storyteller must address. To deliver those goods to the viewer, the creator first has to work through them herself. A well-crafted presentation, then, becomes a thought-provoking win-win on both sides of the camera.

These skills are considered to be so crucial these days that not only will The Promontory have a media studio in the learning space, but the OECD will be including Media and Artificial Intelligence Literacy in the 2029 PISA assessment. This alone speaks to the need for individual educators and collective learning models to ensure that students have worked up to the ability to interrogate the world around them; not in a hostile way, but with a calm and measured confidence. Isn't this the way most of our film heroes respond to the world and command the lead?

The ideas in this post draw on the essay "Chasing Algorithms: Why the Future of Education Depends on the Stories Machines Can't Tell" by Daniela Kon Lieberberg, CEO of SIMA Academy.