The people-centered value orientation is the shared foundation of science popularization and AI literacy. Science popularization practices dominated by instrumental rationality tend to prioritize operational skills while neglecting value choices, power relations, and social consequences. This creates a structural misalignment in which "instrumental competence precedes value judgment," thereby generating a cycle of vulnerability. A profound ethical shift in the science popularization paradigm is essential to breaking this cycle. As a core element of AI literacy, ethics is important not only because it endows users with cognitive capacities and provides the value foundation for safeguarding human subjectivity, but also because AI, as a non-neutral actor, makes it imperative for ethical norms to respond to the demands of social development. Only by placing ethical judgment at the core of science popularization can technological alienation be curbed and the goal of technology for social good be achieved. Science popularization practice should therefore proceed from an ethical standpoint, embed ethics to open up its practical domains, focus on ethical reconstruction, establish value boundaries for technological practice, and elevate AI literacy from an individual skill set to a socially oriented system of competence, thereby activating the dual dimensions of science popularization..