AI ChatGuard

AI chats are NOT private. And AI prompts are NOT secure, according to two peer-reviewed studies by Stanford and MIT.

Big picture: The moment you press send, whatever data you typed or uploaded into an AI prompt is gone. Plus, hackers do attack AI prompts, which can compromise user data. So be smart: Secure your AI chats like you secure your computer.

AI ChatGuard is a simple tool to use. It discreetly sits in your browser and checks every message, file, and image for sensitive data before it reaches an AI tool — then hands you back a moment of choice: what stays private, and what gets shared.

WHAT IT DOES, AND HOW IT WORKS

Three things, and they're easy: Review, Redact, Revise.

• Review — See what AI ChatGuard finds before anything is sent, sorted by type and by risk.

• Redact — Black out sensitive data in text, documents, images, and files with one-click.

• Revise — Safely reword sensitive data with one-click, so your question still works without giving away the details.

It all happens in about a half-second — we live in the gap between AI speeds and network speeds — so it feels fast, accurate and discreet. Think of it as a digital tripwire. Plus a tiny 'do-over machine' to get things right in real time.

Most important, you're always in control. Step by step. Never the AI tool.

Original: "Avery Chen, 58, at Mercy Valley Hospital, on metformin 1000mg ... "

Protected: "A patient in their late fifties, on standard type 2 diabetes medication ..."

The question survives. Your privacy is protected.

WHY IT'S SO ACCURATE

AI ChatGuard isn't just matching patterns. It uses the same kind of detection primitives used by GitHub, Microsoft, and Splunk. In fact, we also use the same techniques that help guard financial and government secrets — across 50+ broad categories and thousands of sensitive data types.

Simply put, AI ChatGuard protects social Security numbers, names, phone numbers, addresses, API keys, payment cards, medical details, company secrets, and more.

And it's smart about context, not just pattern matching. A card number has to pass a validity check before it counts. A medical code only counts in a medical context. And if you're about to overshare, you get a discreet nudge — "Hmmm, try this safer version". We're not in the lecture business.

WORKS WHERE YOU WORK

AI ChatGuard protects you across ChatGPT, Claude, Gemini, and hundreds of other AI chat surfaces — including new ones as they appear. It also protects sensitive data inside the work apps you already use: Gmail, Google Docs, and Microsoft 365 and more.

BUILT TO FAIL SAFE

If protection can't run for any reason, your files are held, not waved through. We think an AI security tool gets exactly one chance to be wrong. Transparency secures trust.

ROGUE AI, TOO

Beyond your own data, AI ChatGuard also spots the rogue AI reaching for it while you browse and chat: agents, bots, probes, and trackers, including unknown ones as they appear. Three choices, one click: Allow, Warn, or Block.

YOUR PRIVACY

Here's a fair question to ask about a software tool: is the software watching you, too? In this case, no.

Detection runs entirely on your device — a safe sandbox. Your prompts, your messages, your images, and your files are never transmitted to us or anyone else. We don't collect your chat data, and we never sell, rent, or trade any data. Under the hood it's built on an open protocol for AI chat surfaces (PPS) and a patented in-band signaling technology (AINI), so the protection travels with your data. The full details are in the privacy policy, which is worth a read.

FOR POWER USERS AND BUSINESSES

If you want more control, it's there: protection modes (Relaxed, Balanced, Strict), fine-grained categories for personal and work data, and saved profiles. Businesses and organizations can use MSP mode to deploy AI ChatGuard across a whole organization with shared policies.

AI ChatGuard was designed as enterprise-grade technology to protect your data. We focused on people and organizations who want to use AI tools safely.

WHAT IT COSTS

AI ChatGuard is free. If you want the full set of user knobs, advanced protections, and manual controls, there is a Pro upgrade. Organizations also can deploy AI ChatGuard company-wide with an MSP upgrade, which includes a multi-tenant IT dashboard with IT policy controls.

THE BOTTOM LINE

AI tools are useful, but they are not private. Most people overshare, which can be problematic. AI ChatGuard protects your data, but it won't make your data disappear after it is shared with an AI tool. Nothing will. But AI ChatGuard does something genuinely useful: it shows you the sensitive data you're about to give away. Then gives you a simple way to keep it. If you care about data security and data privacy, that's worth having.

ABOUT

AI ChatGuard is an independent product of AINCO Software. Our team includes engineers who've worked at companies like Cisco, Microsoft, and Intel, with 40+ years of combined experience in networks, security, and machine learning. AINCO is its own company — not owned by or affiliated with any other company.

OUR PRIVACY PROMISE

We don't collect your chat texts, chat files, or chat images. Zero, zip, nada.

Full Privacy Policy: https://beta.aincosoftware.com/privacy-ai-chatguard

Install AI ChatGuard for secure AI chats. Protect your secrets and don't give away your knowledge work for free.

AINCO Software — [email protected]