> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.humansecurity.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.humansecurity.com/_mcp/server.

# MediaGuard Overview

MediaGuard is HUMAN's predictive, pre-bid advertising fraud prevention system that predicts, in real-time, whether a specific ad opportunity is likely to be [invalid traffic (IVT)](/advertising/ivt-taxonomy) based on information provided from the OpenRTB bid object. It focuses on identifying illegitimate ad impressions that are not being served to, or viewed by, humans in real-time. You can think of MediaGuard as a predictive model based on reputation data. MediaGuard is deployed in tandem with FraudSensor. The information collected via the FraudSensor tag is used to train the MediaGuard algorithms.

> **Note**
>
> You **must** include [FraudSensor](/advertising/fraudsensor-overview) as a component of your overall MediaGuard integration. When used in tandem, FraudSensor can validate the predictions you receive from MediaGuard. This combination of pre-bid and post-bid data also helps us fine-tune MediaGuard's prediction models against new and evolving threats.
>
> To learn more about using MediaGuard and FraudSensor together, see [Closing the Loop](/advertising/closing-the-loop-mediaguard-and-fraudsensor).

## How MediaGuard works

MediaGuard is an intermediary service that sits between the bid requests you receive and the bids that you actually make. After you've set up a [MediaGuard integration](/advertising/mediaguard-integration-options-deployment), you'll be able to send data to MediaGuard and receive real-time predictions about your bid requests.

Every time you receive a bid request, you'll either need to extract certain details from that bid request and send them to MediaGuard or send the bid request's entire OpenRTB `BidRequest` object to MediaGuard. MediaGuard analyzes these details to determine whether the user driving that bid request is valid or invalid. Any invalid bid requests are sorted into different [IVT categories](/advertising/ivt-taxonomy) with information about the specific traits displayed by that bid request.

MediaGuard then delivers bot or fraud decisions on individual ad opportunities in milliseconds. Based on this decision, you can decide whether to continue bidding on a specific ad impression. For example, you might configure your bids to automatically drop any bid requests that receive an `IVT=true` response from MediaGuard. You can also view your MediaGuard data in the [HUMAN Dashboard](/advertising/understand-the-human-dashboard) or via our [Reporting API](/advertising/reference/about-the-reporting-api).

MediaGuard can also return optional Policy decisions in addition to IVT. Policies are opt-in and are not IVT. See [IVT and Policy Reason Codes](/advertising/ivt-taxonomy-reason-codes). For Removed Apps, see [Removed Apps Enforcement](/advertising/removed-apps-enforcement).

## Get started with MediaGuard

To get started with MediaGuard, you'll need to set up an integration with either the [MediaGuard API](/reference/about-the-mediaguard-api) or the MediaGuard SDK.

A HUMAN representative will help you choose the best [integration option](/advertising/mediaguard-integration-options-deployment) to suit your needs, walk you through the [integration process](/advertising/mediaguard-integration-phases), and set up a new [MediaGuard cluster](/advertising/mediaguard-server-clusters) on your behalf.