This pilot study explored the potential of relocating English as a Foreign Language (EFL) reading activities to a café setting supported by AI tools. Sixteen intermediate learners (B1–B2) participated in an eight-week, single-group program that combined Google Lens and ChatGPT with peer discussion and instructor guidance. Data were collected through pre/post self-report surveys and guided post-session conversations documented in field notes. The pre-session survey measured enjoyment of reading in English (an affective-motivational construct), whereas the post-session survey measured comfort in the café setting (a situational-environmental construct); these are reported as independent descriptive profiles rather than as a pre-to-post change score. Findings indicated that learners reported high situational comfort in the café environment, sustained engagement across all sessions, and practical benefits from using AI tools for vocabulary support and text clarification. Usability ratings for the tools were moderate, with some participants citing challenges linked to app-switching and cognitive load; two students expressed reservations about environmental distractions and AI tool complexity. The study demonstrates the feasibility and affective appeal of integrating “third-place” learning environments with AI-assisted reading, while acknowledging limitations of sample size, the absence of a control group, reliance on self-report, and the lack of objective comprehension measures. Because the design cannot rule out novelty effects, Hawthorne effects, or social-desirability bias, results should be interpreted as exploratory. These findings suggest that café-based, AI-supported models may enhance learner engagement and merit further controlled research using waitlist-control or crossover designs.