AI Transparency
How TrainRun uses AI, its limitations, and our commitments to you.
Where AI Is Used
TrainRun uses AI (powered by OpenAI) in the following areas: processing transcripts into structured learning content, generating retrieval quiz questions from your study material, creating module summaries and concept maps, and identifying key vocabulary and concepts. All AI processing happens on our servers - no AI models run on your device.
What AI Does Not Do
AI does not make decisions about your account, subscription, or access. AI does not grade or assess you - quiz scores simply reflect how many answers matched the generated answer key. AI does not store or learn from your personal information. No human reviews your content unless you submit a report.
Limitations
AI-generated content may contain inaccuracies, omissions, or misinterpretations of the source material. Quiz questions are generated automatically and may not perfectly represent the original content. AI may occasionally produce duplicate, ambiguous, or poorly worded questions. The quality of generated content depends on the quality and clarity of the source material.
Accuracy
We design our prompts and validation pipeline to maximize accuracy, but we cannot guarantee that every generated question or summary is correct. TrainRun is a learning aid - not an authoritative source. Always refer to your original study materials for verification.
Your Responsibilities
You are responsible for verifying the accuracy of AI-generated content against your source material. You should not rely solely on TrainRun quizzes for exam preparation, professional certification, or any high-stakes assessment. If you notice an inaccurate or problematic question, please use the Report button available on every quiz.
Our Principles
Transparency: we clearly disclose where AI is used and what its limitations are. Privacy: your content is processed only to generate learning materials and is not used for model training. Educational focus: AI is used solely to support learning through evidence-based techniques like retrieval practice. Continuous improvement: we review reported questions and refine our generation pipeline. Bias reduction: we design prompts to produce neutral, factual content drawn from your source material.
Auditing & Versioning
We store metadata for each generated question set, including the AI model used, prompt version, and generation timestamp. This allows us to audit and reproduce results if a question is reported. When we update our generation pipeline, existing questions are not automatically regenerated - you can refresh a module to get updated content.
Contact
If you have questions about how TrainRun uses AI, contact us at admin@trainrun.io.