| title | date |
|---|---|
Problem Statement |
2025-05-08 |
⚠️ Deadline for submission: 10 July 2025⚠️ Presentation: 18 - 25 July 2025
What is the main functionality ?
Automatic anonymization and summarization of user-submitted documents - PDFs at first, removing personal data such as names and addresses to ensure compliance with legal and privacy requirements.
Users simply upload a document and can either manually mark sections for redaction or let the AI analyze the text and suggest replacements, ensuring fast, consistent anonymization without leaking any real-world details.
Lower Priority:
Voice based interaction with the documents.
Who are the intended users?
- Governments
- Universities
- Law firms
- Any organization handling large volumes of sensitive documents
How do we use GenAI meaningfully?
- Auto-Anonymization
- Automatically detect and redact or replace personal/sensitive information.
- AI-Powered Summaries
- Generate concise summaries to reduce manual reading effort and save users time in reviewing lengthy texts.
- User uploads a file in one of:
PDF,DOCX,TXT.
- The AI scans for personal information (names, addresses, phone numbers, etc.) and redacts or replaces it with placeholders.
- The user can select additional fields which were not auto-detected.
- Users can also edit or rename placeholders (e.g. change
“John Doe”→“Person A”).
- The user can request a summary of the document, which the AI generates based on the content.
- User downloads the anonymized and/or summarized version in their desired format.
- As a user, I want to upload documents in various formats (PDF at first) so that I can work with the files I already have.
- As a user, I want to anonymize my document so that I can share it without revealing sensitive information.
- As a user, I want to manually review the anonymization process so that I can ensure all sensitive information is properly handled.
- As a user, I want to be able to select the level of anonymization (e.g. light, medium, heavy) based on my needs.
- As a user, I want to save and download the anonymized document in my desired format so that I can use it as needed.
- As a user, I want to be able to generate a summary of the document so that I can quickly understand its content.
- As a user, I want an archive of my documents so that I can keep track of my Original vs. anonymized files as well as the generated summary.
(DO NOT include the following user stories in the final document)
- As a user, I want to be able to chat with the document so that I can ask questions and get answers in real-time -> Chatbot
- As a user, I want to be able to talk to the document so that I can interact with it in a more natural way -> Voice based interaction
- As a user, I want to be able to ask questions about the document so that I can get specific information without reading the entire text.
- As a user, I want the system to be able to highlight sections of the document based on the agent conversation so that I can mark important information for later reference (Tool calling).
- As a user, I want to be able to add comments to the document so that I can provide feedback or notes for myself or others.
As a user, I want to upload documents in various formats (PDF at first) so that I can work with the files I already have.
- I want to be able to click a button to select files from my computer.
- I want to be able to drag and drop files into the upload area.
- I want to see a progress bar/loading indicator indicating the upload status.
As a user, I want to anonymize my document so that I can share it without revealing sensitive information.
- I want the system to automatically detect and redact personal information (e.g. names, addresses) from the document.
- I want to see a preview of the anonymized document before finalizing the process.
- I want to be able to select the level of anonymization (e.g. light, medium, heavy) based on my needs.
As a user, I want to manually review the anonymization process so that I can ensure all sensitive information is properly handled.
- I want to see sections of the document that have been anonymized - automatically highlighted.
- I want to see the original text on top of the anonymized text.
- The original text should be red and the anonymized text should be green.
- I want to be able to click on the highlighted sections to edit the anonymized text.
- I want to be able to have freedom to select any other text in the document to be anonymized.
As a user, I want to be able to select the level of anonymization (e.g. light, medium, heavy) based on my needs.
- I want to see a slider or dropdown menu to select the level of anonymization.
- I want to see a description of what each level of anonymization means (e.g. light = names only, medium = names and addresses, heavy = all personal information).
- I want to be able to change the level of anonymization at any time during the process.
- I want to see a preview of the document with the selected level of anonymization applied.
(Optional)
- I want to be able to compare the original document with the anonymized document side by side.
- I want to be able to see the differences between the original and anonymized documents.
As a user, I want to be able to save and download the anonymized document in my desired format so that I can use it as needed.
- I want to be able to click a button to save the anonymized document - Local storage.
- I want to be able to download the anonymized document in my desired format.
- I want to see a confirmation message after the document is saved/downloaded.
As a user, I want to be able to generate a summary of the document so that I can quickly understand its content.
- I want to be able to click a button to generate a summary of the original uploaded document (without having to anonymize it).
- I want to the summarize button to work exactly the same even after the anonymization process is complete.
- I want to be able to save the generated summary in my desired format and download it.
- I want to see a confirmation message after the summary is saved/downloaded.
As a user, I want to have an archive of my documents so that I can keep track of my Original vs. anonymized files as well as the generated summary.
- I want to see a list of all my uploaded documents in the archive.
- I want to have a side by side comparison view to see the original and anonymized versions of each document in the archive (separate page).
- I want to be able to search for specific documents in the archive, based on their file names.
| ID | Title | Acceptance Criteria |
|---|---|---|
| US-1 | Upload Document | - Click to select files from computer - Drag & drop into upload area - Show upload progress bar |
| US-2 | Auto-Anonymization | - Automatically detect & redact PII (names, addresses, etc.) - Show preview of redactions - Offer light/medium/heavy levels |
| US-3 | Manual Review & Customization | - Highlight auto-anonymized sections - Show original text (in red) over anonymized text (in green) - Allow click-to-edit anonymized text - Allow selecting any text to anonymize |
| US-4 | Select Anonymization Level | - Slider or dropdown to choose level - Descriptions for light/medium/heavy - Live preview updates on change |
| US-5 | Save & Download | - “Save” button stores anonymized document locally - “Download” button exports in chosen format (PDF, DOCX, TXT) - Confirmation message on success |
| US-6 | Generate Summary | - “Summarize” button works before or after anonymization - Displays summary of original document - Save/download summary in chosen format - Confirmation message on success |
| US-7 | Document Archive | - List all uploaded documents (original, anonymized, summary) - Side-by-side view of original vs. anonymized - Search by file name in archive |