Abstract:

Secondary schools are increasingly expected to build entrepreneurial competences (initiative, creativity, collaboration, ethical judgment, and financial/digital literacy), while protecting learner wellbeing, equity, and academic integrity. At the same time, Artificial Intelligence (AI) tools (especially generative AI and analytics-enabled learning platforms) are entering classrooms quickly, often as add-ons rather than as elements of purposeful instructional design. This paper presents a tight, practice-oriented framework for integrating AI into secondary entrepreneurship education to strengthen authentic venture learning, student agency, and competency-based assessment, without turning entrepreneurship into automated business-plan production.

The paper proposes the AI-Enabled School Venture Learning Loop (ASVLL), a project-based instructional model applicable to lower and upper secondary education. ASVLL structures entrepreneurial learning into four iterative stages:
(1) problem finding and opportunity exploration,
(2) rapid solution sketching and prototyping,
(3) school-safe validation, and
(4) venture storytelling and basic planning.

Each stage embeds bounded AI supports aligned with learning outcomes. Generative AI is used for divergent ideation, drafting interview questions, and generating alternative value propositions; conversational agents scaffold reflection through structured “claim–evidence–reasoning” prompts; and lightweight learning analytics indicators provide formative signals on progress (e.g., number of testable hypotheses, iteration frequency, stakeholder breadth, and evidence quality). AI is framed as a scaffold rather than an authority: students must log prompts, assumptions, and evidence that confirms or contradicts AI suggestions, creating an auditable learning portfolio.

To ensure feasibility in typical school constraints, the paper outlines a semester-compatible implementation blueprint:
(a) define a small set of observable competences (opportunity recognition, experimentation, collaboration, communication, ethical awareness),
(b) deliver learning through short entrepreneurship “sprints” with clear milestones,
(c) use low-risk validation methods appropriate for minors (peer/community surveys, simulated customer panels, prototype demonstrations, and A/B tests on mock landing pages), and
(d) assess learning through portfolio evidence, team retrospectives, and rubric-based teacher judgment rather than polished AI-generated artifacts.

The framework provides reusable tools, including competence rubrics, reflection templates, peer-feedback protocols, and simple indicators that help teachers detect stalled teams, superficial validation, or over-reliance on AI text generation.

Ethics and integrity are treated as design requirements: transparent AI-use disclosure, prompt/output traceability, explicit instruction on bias and hallucinations, and strict boundaries for data collection during validation (consent, minimal data, age-appropriate engagement). Equity is supported through structured scaffolds for students with diverse language and digital skills, positioning AI as an accessibility enabler rather than a widening force. The paper contributes an adoptable instructional architecture that translates experiential entrepreneurship pedagogy into an AI-ready model for secondary schools, while preserving teacher judgment and student ownership of learning.

Keywords:

Entrepreneurship education, Generative AI, Learning analytics, Competency-based assessment.