The Trust Layer
Between You and AI
InfiniUm Shield automatically masks your personal data, passwords, and API keys before they reach ChatGPT, Claude, or Gemini — without changing how you work.
How It Works
Shield works silently in the background. You use AI exactly as you normally would — Shield handles everything else.
john@company.com becomes [EMAIL_1] before leaving your browser.See It In Action
This is what happens when you send a prompt containing your email and a password.
My email is john@acme-corp.com
and our staging DB password is
Depl0y$ecure2024!
Can you help me fix this SQL error?
masks it
My email is [EMAIL_1]
and our staging DB password is
[PASSWORD_1]
Can you help me fix this SQL error?
✅ The AI still understands the full context and gives a useful answer — but never sees your real data.
Who Needs AI Privacy Protection?
Anyone who uses AI tools in their daily work — which is almost everyone now.
Everything Shield Does For You
Powerful protection that runs invisibly — you won't even notice it's there until it saves you.
What Shield Detects
19 detection rules across 3 severity levels — from critical credentials to personal identifiers.
🔴 Critical — immediate security risk | 🟡 Warning — personal data | 🔵 Info — infrastructure details
Simple, Honest Pricing
The free version gives you real, complete protection. Pro adds team features and advanced capabilities.
- Works on ChatGPT, Claude & Gemini
- All 19 PII detection rules
- Custom masking rules (up to 10)
- Automatic response de-masking
- Per-site on/off toggle
- Masking statistics & history
- Team rule sync
- Audit log
- Advanced detection patterns
- Everything in Free
- Sync rules across all devices
- Team shared dictionary
- Full audit log — what was masked & when
- Advanced detection (env vars, code secrets)
- Unlimited custom rules
- Priority support
Get Protected in 60 Seconds
Installing Shield takes less than a minute. No signup, no configuration, no learning curve.
Not ready for the extension yet?
Try our free AI Prompt Privacy Checker — paste any prompt and instantly see what sensitive data it contains, with a redacted version you can safely send.
Try the free web scannerFrequently Asked Questions
Shield scans your prompt text locally in your browser to detect sensitive data. Nothing — not your prompts, not your personal data — is ever sent to InfiniUm servers. All scanning and masking happens 100% on your device. You can turn off your internet connection and Shield will still work perfectly.
Yes — completely. Shield replaces sensitive values with tokens like [EMAIL_1] or [API_KEY_1]. The AI receives your full question with placeholders instead of real values and can still provide accurate, complete help. The context of your question is preserved. After the AI responds, Shield automatically swaps tokens back to your real values so you read a natural response.
Most commercial AI providers store your prompts for safety review and may use them to improve their models. Their Terms of Service often permit sharing data with third-party partners for marketing purposes. This means emails, passwords, client names, API keys, and any other data you paste into a prompt can be stored on external servers indefinitely — and in some cases legally shared with others. Shield ensures your sensitive data never leaves your browser.
No account needed for the free version. Install the extension and it works immediately — all settings are stored locally in your browser. An InfiniUm account is only required for the upcoming Pro plan features like syncing custom rules across multiple devices.
No measurable impact. The scanning and masking process takes less than a millisecond on any modern device. Shield uses efficient regex patterns and runs entirely in memory — no disk access, no network requests, no waiting. You will not notice any difference in your browsing speed or AI response times.
If you send a prompt with real credentials, treat them as potentially compromised and rotate them immediately — change passwords, regenerate API keys, and update any affected services. This is true regardless of Shield. For credentials, the safest approach is always to assume they may have been logged the moment they leave your device.
The PII detection engine (anonymizer.js) is open and auditable — you can inspect exactly what patterns we detect and how. We believe privacy tools must be transparent. The full extension source will be published on GitHub shortly after the Chrome Web Store launch.
Yes. Shield works on both free and paid accounts for all supported AI tools. Whether you use the free ChatGPT, Claude Free, or Google Gemini, Shield protects your prompts identically. No subscription to any AI service is required.
Start protecting your prompts today
Install InfiniUm Shield in under 60 seconds. Your personal data, passwords, and API keys stay yours — always.