CID222 builds a unified AI security platform that sits between your applications and LLM providers, protecting sensitive data across your AI infrastructure โ enterprise-level privacy, compliance, visibility and control.
- ๐ PII detection & masking โ 15+ entity types caught before they reach the LLM
- โฃ๏ธ Toxicity filtering โ hate speech, abuse and profanity blocked in 100+ languages
- ๐ Prompt injection & jailbreak prevention
- ๐จโ๐ป Code safety โ malicious / unsafe code flagged in prompts and outputs
- ๐ Document & image redaction โ OCR-powered PII removal from PDFs, DOCX and images
- ๐ Compliance-ready โ GDPR, EU AI Act, KVKK; full audit trails
๐ก๏ธ CID Guardrails
This organization hosts the CID Guardrails model suite โ multilingual transformer classifiers that power the platform's input/output filtering pipeline. Models are fine-tuned on our in-house multilingual datasets (also hosted here) and exported to ONNX for low-latency production inference.
| Guardrail | Task | Base model |
|---|---|---|
| ๐ PII / NER | Token classification | DistilBERT (multilingual) |
| โฃ๏ธ Toxicity | Text classification | mDeBERTa (multilingual) |
| ๐ Prompt Injection | Text classification | mDeBERTa (multilingual) |
| ๐จโ๐ป Code Safety | Text classification | UniXcoder |
Browse our collections for the latest model versions and their training datasets.