AI literacy as the slowest but most durable layer of defense, moving users across typology profiles¶

Strategy · All strategies
Evidence · 1 claim (1 for) · 2 studies (2 associational), q2 · 1 of 2 report an effect size

Description¶

The article closes by positioning AI literacy differently from the other interventions: it "operates differently by gradually increasing users’ capacity to verify and potentially moving them across profiles." Literacy gives users a working model of what a language model is and why it can fabricate, which turns vague unease into targeted checking. Within the Swiss cheese frame it is the layer that closes gaps slowly but durably, complementing affordances and engagement mechanisms.

Design Implications¶

Context¶

Requirements¶

• Literacy building requires time and accumulated experience, since expertise and AI literacy accumulate through experience

Constraints¶

• It is the slowest layer of defense, acting gradually rather than within a session

Target Learners¶

• Chatbot users lacking domain expertise, AI literacy, or both

Target Learning Goals¶

• Building a working model of language models and targeted verification habits

Affordances¶

• Swiss Cheese Layered Trust Defense
• Ability Motivation Verify Typology

Claims¶

• Profiles differ in immediate usage behaviors: Engaged and Literate users verify most, while Skeptical Minimal users report the highest shortcut reliance [+M]

Related Strategies¶

• Six engagement-based incentive mechanisms that build evaluative moments into chatbot interaction

Examples¶

-

Key Sources¶

• Jaidka K and Cai M. (2026). Why we believe chatbots: trust calibration as a design problem. Front. Psychol. 17:1935527. https://doi.org/10.3389/fpsyg.2026.1935527