From the motivation section:

Too many hypervisor projects out there are either extremely complicated (Xen, KVM, VirtualBox) and/or closed-source (VMware, Hyper-V), as well as heavily focused toward Linux-based development or system. Additionally, most (other than Hyper-V) of them are expressly built for the purpose of enabling the execution of virtual machines, and not the virtualization of a live, running system, in order to perform introspection or other security-related tasks on it.

What exactly falls under the "other security-related tasks" section? The only thing I can think of would be for a privileged application to try and hide things from the OS.

Apart from being able to hide from the OS a hypervisor is also safe from manipulation by the OS and can observe and manipulate anything the OS does.

This opens a few doors: debugging or reverse engineering the OS or kernel modules, sandboxing a system you don't trust, detecting rootkits, protecting your antivirus solution from malware, etc.

You could use it to load an image abd inspect how it behaves maybe ?

There is a comprehensive "motivation" section on the linked page of the project.