Skip to main content
🌍 AI for Everyone

Privacy & Staying Safe

AI chats are incredibly useful, but you're typing into a service run by a company. A little awareness keeps you safe without making you paranoid.

Assume what you type may be stored β€” and sometimes reviewed or used to improve models. Many services let you turn off training on your chats or delete history, but the safe default is: don't paste anything you'd be upset to see leaked.

Be careful with:

  • Personal identifiers β€” full national ID numbers, passwords, bank/card numbers.
  • Other people's private info β€” medical details, a friend's address, work secrets covered by an NDA.
  • Confidential work data β€” many jobs forbid pasting internal documents into public AI tools. Check your employer's policy.

Safe and smart habits:

  • Redact before you paste β€” swap real names/numbers for placeholders like "[NAME]" or "[ACCOUNT]". The AI can still help with the structure.
  • Use privacy settings β€” turn off chat-history training if you prefer; delete conversations you don't need.
  • Be skeptical of "AI" links and apps β€” scammers clone popular tools. Use the official website or app store listing.

You stay in control. AI is a tool you point at a problem β€” not a confessional.

What's the safest default assumption about what you type into an AI chat?

You want AI's help rewriting a letter that contains a real account number. Best move?

Before pasting an internal company document into a public AI tool, you should…