“Vibe‑coding” emerged in 2023 as a social‑media label for an intuitive style of programming in which developers feed natural‑language prompts to generative AI and accept whatever “feels right” in the returned code. Within two years, the jargon migrated into education, spawning phrases such as '“vibe‑teaching” and “vibe‑learning”. Posts and conference talks now describe teachers who “vibe” their way through lesson preparation, letting AI produce slides, quizzes, or even feedback, while students “vibe‑learn” by asking chatbots to assemble essays or explanations on demand. Although the imagery is catchy, this paper argues that the vibe metaphor is pedagogically misleading.

First, vibe denotes a non‑directed, affect‑driven search. Effective teaching and learning, by contrast, are intentional, goal‑oriented, and reflexive processes that hinge on clear objectives, scaffolding, and iterative feedback. Treating AI‑assisted pedagogy as merely “going with the vibe” risks normalizing superficial engagement, eroding professional judgment, and masking the labour of curating, verifying, and contextualizing machine‑generated material.

To reclaim conceptual precision, we revisit Csíkszentmihályi’s theory of flow, a state of deep concentration that arises when individuals tackle optimally challenging tasks supported by immediate feedback. Translating this construct to AI‑enhanced pedagogy, we introduce the operational term “flow‑teaching”: the deliberate orchestration of human–AI interaction so that both teacher and learners enter a flow state in pursuit of explicit learning goals. In flow‑teaching, AI tools serve as responsive collaborators, generating exemplars, modeling thought processes, or adapting explanations, while the educator sets the trajectory, calibrates difficulty, and embeds moments for reflection.

Framing practice around flow corrects three shortcomings of the vibe metaphor. It:
(1) foregrounds purposeful design over serendipity,
(2) situates AI within an evidence‑based account of motivation and expertise development, and
(3) provides axioms for evaluating when AI augments rather than displaces human agency.

The paper maps a brief genealogy of vibe‑coding, tracks its diffusion into educational discourse, and then synthesizes psychological, instructional‑design, and human–AI collaboration literature to articulate principles of flow‑teaching.

The contribution is conceptual and twofold: it critiques a rapidly popularising yet under‑examined metaphor, and it offers a theoretically grounded alternative that aligns emerging AI capabilities with foundational insights into effective pedagogy. By replacing vibes with flow, we invite researchers, teacher‑educators, and higher‑education practitioners to develop, test, and refine design patterns that keep human intention, expertise, and reflection at the centre of AI‑rich classrooms.

Keywords: Flow‑Teaching, Flow‑Learning, AI‑Enhanced Pedagogy, Human–AI Collaboration, Instructional Flow.