30 Days Gen AI Risk Trial -Start Now
Skip to main content
Aona
Book a demo

AI DATA LOSS PREVENTION

Keep sensitive work out of AI submissions.

Protect sensitive prompts and uploads with policy-driven redaction and blocking. Give employees a safer way to work with AI, and security a control point before submission.

Start free trial Book a demo
Find the right product for your workflow

For supported workforce AI interactions. Test the exact client and input path in scope.

Redacted

Sensitive fields replaced

The request can continue with customer and account identifiers removed.

[CUSTOMER][ACCOUNT]

Draft a renewal note for [CUSTOMER] · account [ACCOUNT].

Synthetic example · No prompt is sent.

BUILT FOR EMPLOYEE AI USE

The AI tools your teams already use.

OpenAI logoOpenAIDLP for ChatGPTMicrosoft logoMicrosoftDLP for Microsoft CopilotGoogle logoGoogleDLP for Google GeminiAnthropic logoAnthropicDLP for Claude

Aona tracks 10,000+ AI tools in its risk catalogue. That catalogue breadth is not a claim of identical enforcement for every tool.

BEFORE THE PROMPT IS SENT

Put your data policy where the work happens.

Move from a written rule to an employee-facing response on the supported AI workflow.

  1. 01

    Inspect the input

    Detect sensitive entities in prompts and supported uploads through the configured browser or endpoint path.

  2. 02

    Apply your policy

    Use the data category, AI tool and employee group to determine the supported redaction or blocking response.

  3. 03

    Guide the employee

    Make the response visible. Show what was protected, or explain why the submission was blocked.

Renewal brief

Customer[CUSTOMER]
Account[ACCOUNT]
Renewal value[FINANCIAL]
Synthetic document output · Fictional sample

DOCX · XLSX · PDF

Protect the file.
Keep the format.

Replace sensitive entities without turning a useful document into an unreadable one. Aona’s file-redaction workflows preserve layout for supported DOCX, XLSX and PDF uploads.

Use the file-redaction walkthrough to check the document type, upload path and output your team needs.

Review file-redaction workflows

Works alongside existing DLP

AI DLP is not a replacement project.

Your existing controls remain responsible for their established email, file-share and endpoint use cases. This evaluation focuses on sensitive prompts and uploads at supported AI interactions.

What to record in a repeatable AI DLP test
  1. Environment: AI app, account type, operating system, browser or native client, and installed Aona version.
  2. Input: synthetic data category, typed or pasted prompt, or the exact file and upload method.
  3. Expected response: the configured blocking or redaction rule, plus an allowed input as a control.
  4. Observed result: attempted submission, employee notice, protected output and matching event, where available.

Keep observed results blank until the installed test is run. Repeat separately for another client, input method or file format.

Download a blank test record
Review policy evaluation

Review source code, secrets and AI debugging paths

Buying questions

AI data loss prevention questions

What is AI data loss prevention?
AI data loss prevention evaluates and controls sensitive prompts and uploads before they reach a supported AI service. A policy can block a submission or redact sensitive entities, depending on the configured workflow.
Does Aona support ChatGPT, Copilot, Gemini and Claude?
Aona provides data protection for supported ChatGPT, Microsoft Copilot, Google Gemini and Claude interactions. Coverage depends on the browser or native client, provider surface, prompt or upload path, and configured policy. The app-specific guides explain what to check for your workforce’s workflow; a provider name alone is not a promise that every surface has identical controls.
Can Aona block or redact sensitive data?
Aona provides hard-block data loss prevention on supported top-tier AI workflows. For supported document workflows, redaction can replace sensitive entities before upload. Test the intended policy response with synthetic material before rollout.
Does 10,000+ tracked AI tools mean every tool has the same enforcement?
No. The 10,000+ figure describes Aona's tracked AI risk catalogue. It does not mean every tool has the same active policy enforcement as a supported top-tier workflow.
Does AI DLP replace our existing DLP?
No. Existing DLP may continue to govern email, file shares and established endpoint controls. This evaluation addresses sensitive prompts and uploads in supported AI interactions.
Can Aona redact DOCX, Excel and PDF files before AI upload?
Yes. Aona supports layout-preserving redaction for DOCX, XLSX and PDF files on supported upload workflows. Sensitive entities are replaced according to policy while keeping the document usable. Confirm the application, file type, upload path and expected output with representative synthetic files before deployment.
Where are prompts processed, and where is the backend hosted?
These are separate deployment choices. Aona’s backend can run in the customer’s cloud, on-premises or on Aona-managed servers. Prompt processing can run on the user’s device (edge), in the customer’s cloud or on-premises environment, or on Aona-managed servers. Confirm the supported configuration for your required features. These choices do not determine where a third-party AI provider processes data.

Focused evaluation

Choose one input. Test the protection it needs.

We’ll map the app, managed client, data category and policy outcome that matter to your rollout.

Book a demo