Trust & Safety

The Scholar Learning Navigation System uses current work, recent attempts, and bounded session context to choose a useful next kind of support. It can misread the situation, choose an unhelpful approach, or produce an incorrect explanation. It should be treated as learning support, not as a perfect diagnosis, mastery measure, or guaranteed source of truth.

Scholar combines teacher-designed learning guidance with AI. It is designed to use explanations, examples, questions, hints, strategy, guided steps, and supported representations while leaving meaningful thinking for the student to do. Scholar+ is designed to adapt the learning approach when the current one is not helping rather than simply repeat the same answer.

Important conclusions should still be checked against class materials, a teacher, or another trusted source, and students should follow their teacher's rules for AI use.

Teacher is a separate human support role, not another AI character and not an automatic check on every AI response. Students review a Teacher handoff before submitting it. Teacher support uses direct Teacher Questions and depends on eligibility, educator assignment, capacity, and educator availability. Responses are asynchronous rather than live chat.

Scholar+ includes 8 direct Teacher Questions per monthly cycle when new memberships become available. Included Questions do not roll over. Teacher responses remain subject to eligibility, capacity, and educator availability.

Signed-in Learning Sessions can retain conversation, student work, recent approaches, supported Learning Objects, and related Teacher feedback for continuity. The learning system also uses bounded derived session context so Scholar can work from the same learning situation without forcing the student to start over.

Current implementation does not automatically build a permanent inferred learner profile across sessions or define a student as a fixed type of learner. Visible history, uploads, Teacher records, and temporary derived learning state have separate data-handling and retention rules.

Student text sent to AI learning routes is screened before model generation in the supported paths. The app blocks categories of unsafe or highly sensitive requests and can show crisis resources where appropriate. These controls reduce risk but do not guarantee that every unsafe or incorrect output will be prevented.

Library is private to the individual student experience. Students do not see other students' responses, and there are no public followers, popularity counts, leaderboards, or public posting mechanics. Authorized staff access remains governed by the application's role and assignment rules.

Scholar+ is designed around Universal Design for Learning ideas, including supported representations, readable layouts, keyboard-aware controls, assistive-technology labels, large-text/reflow behavior, and reduced-motion options in supported views. Accessibility is an ongoing QA area rather than a claim that every screen has completed a formal WCAG audit.

• Self-service access uses an age-range eligibility check before account information is collected. The current self-service experience is for eligible users age 13 or older.
• Signed-in Learning Sessions, messages, submitted work, and Learning Objects can be saved for continuity. AI requests are also processed by the configured AI service.
• Teacher-request records and related uploads are private application data under the product's access rules.
• Saving or uploading a document does not guarantee that AI has read or understood it.
• Scholar+ does not sell personal information or run third-party advertising.

Full detail is in the Privacy Policy.

Questions about access, correction, or deletion can be sent to studio404media@gmail.com.