What is "X-Content-Type-Options=nosniff"?

It prevents the browser from doing MIME-type sniffing. Most browsers are now respecting this header, including Chrome/Chromium, Edge, IE >= 8.0, Firefox >= 50 and Opera >= 13. See :

https://blogs.msdn.com/b/ie/archive/2008/09/02/ie8-security-part-vi-beta-2-update.aspx?Redirected=true

Sending the new X-Content-Type-Options response header with the value nosniff will prevent Internet Explorer from MIME-sniffing a response away from the declared content-type.

EDIT:

Oh and, that's an HTTP header, not a HTML meta tag option.

See also : http://msdn.microsoft.com/en-us/library/ie/gg622941(v=vs.85).aspx

Description

Setting a server's X-Content-Type-Options HTTP response header to nosniff instructs browsers to disable content or MIME sniffing which is used to override response Content-Type headers to guess and process the data using an implicit content type. While this can be convenient in some scenarios, it can also lead to some attacks listed below. Configuring your server to return the X-Content-Type-Options HTTP response header set to nosniff will instruct browsers that support MIME sniffing to use the server-provided Content-Type and not interpret the content as a different content type.

Browser Support

The X-Content-Type-Options HTTP response header is supported in Chrome, Firefox and Edge as well as other browsers. The latest browser support is available on the Mozilla Developer Network (MDN) Browser Compatibility Table for X-Content-Type-Options:

https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/X-Content-Type-Options

Attacks Countered

1. MIME Confusion Attack enables attacks via user generated content sites by allowing users uploading malicious code that is then executed by browsers which will interpret the files using alternate content types, e.g. implicit application/javascript vs. explicit text/plain. This can result in a "drive-by download" attack which is a common attack vector for phishing. Sites that host user generated content should use this header to protect their users. This is mentioned by VeraCode and OWASP which says the following:

   This reduces exposure to drive-by download attacks and sites serving user uploaded content that, by clever naming, could be treated by MSIE as executable or dynamic HTML files.

2. Unauthorized Hotlinking can also be enabled by Content-Type sniffing. By hotlinking to sites with resources for one purpose, e.g. viewing, apps can rely on content-type sniffing and generate a lot of traffic on sites for another purpose where it may be against their terms of service, e.g. GitHub displays JavaScript code for viewing, but not for execution:

   Some pesky non-human users (namely computers) have taken to "hotlinking" assets via the raw view feature -- using the raw URL as the src for a <script> or <img> tag. The problem is that these are not static assets. The raw file view, like any other view in a Rails app, must be rendered before being returned to the user. This quickly adds up to a big toll on performance. In the past we've been forced to block popular content served this way because it put excessive strain on our servers.