Luckily, the vulnerability was already fixed by Apple
The vulnerability (CVE-2016–4631) resides in ImageIO, an API used to handle image data, and works across all widely-used Apple operating systems:
When rendered by applications that use the Image I/O API, a specially crafted TIFF image file can be used to create a heap based buffer overflow and ultimately achieve remote code execution on vulnerable systems and devices.
So, all an attacker needs to do is create an exploit for the bug and send it via a multimedia message (MMS) or iMessage inside a crafted TIFF file:
This vulnerability is especially concerning as it can be triggered in any application that makes use of the Apple Image I/O API when rendering tiled TIFF images. This means that an attacker could deliver a payload that successfully exploits this vulnerability using a wide range of potential attack vectors including iMessages, malicious web pages, MMS messages, or other malicious file attachments opened by any application that makes use of the Apple Image I/O API for rendering these types of files.
Furthermore, depending on the delivery method chosen by an attacker, this vulnerability is potentially exploitable through methods that do not require explicit user interaction since many applications (i.e. iMessage) automatically attempt to render images when they are received in their default configurations. As this vulnerability affects both OS X 10.11.5 and iOS 9.3.2 and is believed to be present in all previous versions, the number of affected devices is significant.
Apple has patched this critical issue in iOS version 9.3.3: then update ASAP!