Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,63 @@
|
|
|
|
|
| 1 |
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: static
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
```markdown
|
| 2 |
---
|
| 3 |
+
title: DataAnonymiser
|
| 4 |
+
emoji: 🛡️
|
| 5 |
+
colorFrom: blue
|
| 6 |
+
colorTo: purple
|
| 7 |
sdk: static
|
| 8 |
pinned: false
|
| 9 |
---
|
| 10 |
|
| 11 |
+
# DataAnonymiser
|
| 12 |
+
|
| 13 |
+
**Local-first PII detection and anonymisation for AI workflows.**
|
| 14 |
+
|
| 15 |
+
DataAnonymiser helps you detect and anonymise sensitive information before documents and data are shared with AI systems.
|
| 16 |
+
|
| 17 |
+
🌐 **[dataanonymiser.com](https://dataanonymiser.com)**
|
| 18 |
+
|
| 19 |
+
## Privacy before AI
|
| 20 |
+
|
| 21 |
+
Using AI with confidential information creates a simple problem: sensitive data may leave your device before it is processed.
|
| 22 |
+
|
| 23 |
+
DataAnonymiser is built around a different workflow:
|
| 24 |
+
|
| 25 |
+
**Original data → Detect PII locally → Anonymise → Use with AI**
|
| 26 |
+
|
| 27 |
+
Personally identifiable information can be removed or replaced before content is shared with services such as ChatGPT, Claude, or other LLMs.
|
| 28 |
+
|
| 29 |
+
## What we're working on
|
| 30 |
+
|
| 31 |
+
Our work focuses on:
|
| 32 |
+
|
| 33 |
+
- PII detection
|
| 34 |
+
- Data anonymisation and pseudonymisation
|
| 35 |
+
- Document and PDF redaction
|
| 36 |
+
- Privacy-preserving AI workflows
|
| 37 |
+
- Local-first processing
|
| 38 |
+
- Sensitive data protection for LLM workflows
|
| 39 |
+
|
| 40 |
+
## Why local-first?
|
| 41 |
+
|
| 42 |
+
Anonymising sensitive information through another cloud service can simply move the privacy problem elsewhere.
|
| 43 |
+
|
| 44 |
+
DataAnonymiser is designed so that sensitive-data detection and anonymisation can happen locally before information is shared with external AI services.
|
| 45 |
+
|
| 46 |
+
## Hugging Face
|
| 47 |
+
|
| 48 |
+
We use this organization to explore and share resources related to:
|
| 49 |
+
|
| 50 |
+
- PII detection models
|
| 51 |
+
- Anonymisation techniques
|
| 52 |
+
- Privacy-focused AI
|
| 53 |
+
- Evaluation and benchmarking
|
| 54 |
+
- Local inference
|
| 55 |
+
- Document privacy
|
| 56 |
+
|
| 57 |
+
More resources and demos will be added over time.
|
| 58 |
+
|
| 59 |
+
## Learn more
|
| 60 |
+
|
| 61 |
+
🛡️ **DataAnonymiser**
|
| 62 |
+
[https://dataanonymiser.com](https://dataanonymiser.com)
|
| 63 |
+
```
|