When AI becomes 'teacher': Is knowledge being 'molded' by algorithms?

In the context of strong technological development, artificial intelligence (AI) is increasingly entering people's learning and research life. From high school classrooms to university lecture halls, from online courses to in-depth research, AI has become the new "teacher", supporting quick and convenient access to knowledge.

AI "teacher" takes the throne

Tools like ChatGPT, Gemini or Copilot... are capable of answering questions in almost every field from foreign languages, programming, science to soft skills in just a few seconds.

AI can not only provide answers, but also explain in detail, provide examples, and even personalize learning paths based on each person’s level and needs. This saves time, reduces costs, and opens up access to knowledge for those who have difficulty with traditional learning methods.

Many teachers and parents see AI as a teaching assistant or home tutor, helping with lesson planning, practice, and reinforcement. Students use AI to research, do homework, and practice skills...

From this perspective, AI seems to be opening a wide door of knowledge, helping learners progress faster in their journey to conquer knowledge.

Ngoc Thien, an information technology student, said he often uses ChatGPT to quickly solve programming exercises. Just enter the request, AI immediately gives complete code with explanations, saving hours compared to searching on Google or flipping through books.

When knowledge is viewed through the "lens" of algorithms

Behind that benefit, however, lies a worrying question: is human knowledge actually expanding, or is it being constrained by the very frameworks that algorithms create?

AI, no matter how smart, is still just a product of training data. Each model is “fed” with a pool of available information that has certain biases, shortcomings, and limitations. The answer that AI provides is not the “absolute truth,” but rather a version of information that has been selected, synthesized, and optimized in a way that the algorithm deems most appropriate, safe, or popular.

According to the MIT Media Lab’s Your Brain on ChatGPT (6-2025) study, the overuse of AI in essay writing results in homogeneous output, with many essays being judged as “soulless” and showing signs of reduced critical thinking. When learning to program, AI also often guides students in the “most common way,” ignoring creative or less common but still effective methods.

As a result, the ability to search for and compare information gradually fades away. Learners easily accept the first answer as the truth, instead of continuing to ask questions or learn more. Over time, thinking is shaped by the algorithm, causing diversity and creativity in accessing knowledge to be narrowed.

Open knowledge or closed knowledge?

AI will certainly continue to play an important role in education , but it cannot and should not completely replace active human learning. Instead of viewing AI as the ultimate “teacher,” we need to see it as a useful but controlled support tool.

Learners must actively verify information, comparing it with a variety of sources, both online and traditional. More importantly, they need to practice critical thinking and the ability to ask deep questions, to not only receive knowledge, but also understand its context, origin and limitations.

Knowledge only truly expands when humans dare to step out of the mold created by algorithms. At that time, AI is not a framework that binds thinking, but becomes an extended arm that helps us go further on the path of discovery and creativity.