answers.delivery

How-to · 3 min read

How to stop staff pasting sensitive data into AI chats

Security leads know staff paste customer files, contracts and code into AI chats, and want to stop sensitive data leaving without banning AI. People are not doing it to cause harm. They want to finish a report faster or fix a bug before lunch. But an email address, a bank number or a client contract pasted into a public chat is still data that has left the company. You want a guard rail that works at the moment someone hits send, not a report that shows up a month later.

Published
10 Nov 2026
Delivery
No. 11
Topic
AI and agents
Evidence
9 claims quoted from milvago.ai

01
Why it's hard

A written AI policy helps, but people forget it the moment they are busy, and nobody reads a policy while drafting a prompt. Blocking every AI site pushes staff to their phones or personal laptops, where you see even less. Network tools inspect traffic in transit, which often means decrypting connections, and anyone working off the VPN or from home may not pass through them at all. By the time a sensitive prompt is spotted in a log, it has already been sent. And a block list goes stale fast, because new AI tools appear every few weeks and the one your team uses tomorrow may not be on it.

This part describes the usual way. It makes no claims about Milvago AI, so nothing is quoted here.

02
How Milvago AI does it

Milvago AI works on the workstation, where the prompt is typed, instead of in the network. A browser extension and a lightweight agent apply your rules on each device, and sensitive data such as email addresses or bank details is masked in the prompt before it is sent to a supported AI service. For each service you choose to observe, block, redirect people to the AI you approved, or mask locally. The rules are signed, they expire, and they keep applying when a laptop is offline, while the usage record stays on your own server. It also records the names of files sent, never the files themselves, and the extension spots visits to more than 160 AI platforms by their address.

Sensitive data is detected and masked on the workstation, before the prompt leaves for a supported AI service.Quoted from milvago.ai
Choose a posture for each service: observe, block, redirect to the AI you approved, or mask locally.Quoted from milvago.ai
Rules are signed, they expire, and they keep applying when a workstation is offline.Quoted from milvago.ai

03
Step by step

  1. 1Install the self-hosted server and console on infrastructure you control.
  2. 2Roll out the browser extension and the agent to the workstations you want to cover.
  3. 3Check the usage map to see which AI services people already reach.
  4. 4Set a posture for each service, such as mask for one tool and redirect for another.
  5. 5Publish the signed policy and review the activity log as events come in.

04
Tips

Start in observe mode for a few weeks. Seeing which tools people actually use tells you where a mask or redirect rule will matter, and where a block would only push them elsewhere.

See which AI services your teams use, and from which workstations.Quoted from milvago.ai

Redirect rather than block when you already pay for an approved AI tool. People keep their habit of asking an assistant, and the data goes to the service you trust.

redirect to the AI you approvedQuoted from milvago.ai

Decide early whether you need prompt text at all. Recording facts about usage, not the content, is often enough for an audit and is easier to defend to staff.

By default an event records facts, not the text of prompts.Quoted from milvago.ai

Tell your teams what the tool does and does not see before you roll it out. Clear rules and an approved option usually cut risky pasting more than surprise monitoring does.

General advice. Not a claim about the product.

05
Questions

Does it decrypt browser traffic?

No. Requests go straight from the browser to the AI service, without TLS interception in the browser.

Requests go straight to the AI service, with no TLS interception in the browser, and the usage record stays on your own server.Quoted from milvago.ai

Is there a free edition?

Yes. Community is free, while Enterprise is licensed per workstation on quote with a 30-day trial.

Download Community, freeQuoted from milvago.ai

Does it replace our network DLP or CASB?

No. It works alongside a network security stack rather than replacing it.

Milvago complements a network security stack; it does not replace it.Quoted from milvago.ai

More from Milvago AI

All 2 answers