ScribeMuni AI Usage & Ethical Charter

At ScribeMuni, we are committed to building a transparent, professional, and secure workspace for civil services preparation. This AI Usage Charter documents how we utilize machine learning, optical character recognition (OCR), and large language models (LLMs) to grade and evaluate your UPSC Mains answers.

1. UPSC Calibration & Grading Methodology

ScribeMuni does not generate arbitrary scores. Our evaluation framework is meticulously calibrated against historical topper answers and the standard evaluation rubrics of the UPSC Civil Services Examination. The AI engine evaluates submissions across five major dimensions:

• Direct Demands & Directives: Inspecting whether the candidate fully answered each keyword command (e.g., "Discuss," "Critically Analyze," "Elaborate," "Evaluate").
• Structure & Balance: Ensuring standard structural flow: Introduction, multi-dimensional body arguments (e.g., PESTLE framework), and a balanced, optimistic, forward-looking Conclusion.
• Content Depth: Searching for crucial structural components like relevant Supreme Court cases, specific Constitutional Articles, Committee Reports (e.g., ARC-II, Ashok Dalwai), and authentic economic stats.
• Value-Additions: Recommending diagram blueprints, schematic charts, and relevant geographic or flow maps to boost presentation marks.
• Administrative Aptitude: Reviewing whether the overall tone is pragmatic, balanced, constitutionally aligned, and suited for a future public servant.

2. OCR Parsing Integrity

When candidates upload handwritten answer sheets, our OCR engine extracts written sentences strictly for context mapping. The digitized transcript is mapped onto the evaluation schema in memory. The uploaded PDF or image file itself is processed under secure, high-integrity transport layers and is permanently dismissed from the operational pipeline immediately after scorecard creation.

3. Intellectual Property Guarantee

We assert no intellectual property rights over any candidate scripts. Every essay draft, mock answer, outline, and structural brainstorming skeleton submitted to ScribeMuni remains 100% your exclusive property. We do not use candidate submissions to create study models, blog materials, or third-party practice books unless a candidate explicitly opts to share their scorecard to the public CSE Peer Review Community feed.

4. Transparency and AI Accountability

We believe AI is a powerful assistant that democratizes high-quality UPSC guidance, which historically has been locked behind expensive coaching networks. However, candidates should use our scorecards as a smart, objective benchmark to accelerate self-assessment and feedback loops. ScribeMuni does not replace real administrative judgment, and we encourage aspirants to actively pair AI insights with peer reviews and professional mentoring.