The recent tragic GermanWings crash has illustrated an important point: even the best designed safety systems can be defeated in scenarios where a trusted individual decides to go rogue.
In the case of the GermanWings crash, the co-pilot was able to lock the pilot out of the cockpit. The cockpit door locking mechanism is designed to enable a trusted individual inside the cockpit from preventing an unwanted person from being able to enter.
Such safeguards exist in security mechanisms in information systems. However, these safeguards only work when those at the controls are competent. If they go rogue, there is little, if anything, that can be done to slow or stop their actions. Any administrator with responsibilities and privileges for maintaining software, operating systems, databases, or networks has near-absolute control over those objects. If they decide to go rogue, at best the security mechanisms will record their malevolent actions, just as the cockpit voice recorder documented the pilot’s attempts to re-enter the cockpit, as well as the co-pilot’s breathing, indicating he was still alive.
Remember that technology – even protective controls – cannot know the intent of the operator. Technology, the amplifier of a person’s will, blindly obeys.