Content Safety Pro
Prompt-injection shields, hallucination grounding, copyrighted-material detection, and multimodal extraction.
Azure AI Content Safety's four flagship features — at a Brainiall price.
What you get
/v1/safety/prompt-shieldPrompt Shields (jailbreak / injection)
Classify any prompt before it reaches your LLM: jailbreak, prompt injection, data exfiltration, impersonation, or none. Returns severity + category + reason + confidence.
Engine: Brainiall Prompt Shield engine
/v1/safety/groundednessGroundedness Detection
Check whether a claim is actually supported by a source text. Returns grounded bool + supporting span + confidence + reason — the building block for hallucination-aware RAG.
Engine: Brainiall Groundedness engine
/v1/safety/protected-materialProtected Material Detection
Detect copyrighted text in user input (famous lyrics, literary openings, proprietary code fragments). Returns matched spans + source attribution. DB grows over time.
Engine: Brainiall Protected Material engine
/v1/content/understandMultimodal Content Understanding
Define a schema of fields you want; send a document image, text, or both. Returns structured JSON for any source modality. The Brainiall answer to Azure AI Content Understanding.
Engine: Brainiall Content Understanding engine
How we compare
| Provider | Equivalent surface |
|---|---|
| Azure AI Content Safety | Has Prompt Shields, Groundedness, Protected Material, Content Understanding — $0.38-1.50/1k records |
| AWS Bedrock Guardrails | Has prompt filtering but no public Groundedness or Protected Material APIs |
| Brainiall | All four endpoints LIVE today, share text-intelligence / document-intelligence quotas, no separate Safety tier. |
Pricing
Content Safety Pro endpoints share your existing Brainiall NLP, Document, and Speech AI usage — no separate bundle subscription. The Free tier covers ~100-1000 calls/month per endpoint.