This vulnerability allows attackers to execute arbitrary code on vulnerable installations of Apple OS X. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file.
The specific flaw exists within the handling of calls to IOHIDSecurePromptClient. The issue lies in the failure to properly sanitize user-supplied pointers before they are dereferenced. An attacker can leverage this vulnerability to overwrite arbitrary kernel memory.