Cyber security is always about people rather than the actual technology. Either people are the threat themselves or those around them, or they constitute a covert insider or outsider threat. On another level some may design design hardware and software that poses a risk.
Stuxnet and Flame are possibly the most sophisticated virus yet to appear on the net. Stuxnet was designed to target Iranian uranium enrichment facilities in that country. This virus targeted one specific set of controllers and contained inside knowledge about operating system. It not only got past a firewall: it actually got into centres disconnected from the network. The suspicion is therefore that it was introduced either maliciously or through carelessness: the use of a shared memory stick or printer allowed it in.
In the long run, the smart grid should be relatively more secure, since the cloud, where much of the processing will take place, is made up of different sized dynamic networks: there is no single operating system or dominant set of protocols, which makes it harder for the “bad boys” to get in and do serious damage.
As analogy: a burglar breaks into a house expecting to have instant access to the homeowner’s valuables. Instead, on coming through the front door, he finds himself in a room with ten identical doors. Going through one of those, he finds another room with ten doors, and so on. The burglar is quickly lost – and isolated.
The “internet of things” has much in common with this architecture: it will be an absolute maze, within which it will be easy to isolate perpetrators and slam the door behind them. In theory, it is not difficult to quarantine attacker software.
Despite this, the rush to get into smart grid is seeing products that are weak, and known to be security weak, on the premise that they will be fixed later. Threats to individual privacy or even the threat from “kiddyhackers” are irritants. However the big threat is terror group wishing to switch off a nation’s power grid.
The real threat is the infiltration of power stations and vital infrastructures. If you wish to attack a country, one of the first tasks must be to bring down the power grid, leading to no transportation, logistics or banking, waster, waste disposal and food – and within a month serious widespread illness. The worry should be that in any cyber war, there would be a takedown of power stations, and this should be recognised as a national security issue.
Despite this, much of the technology that supports telecoms and smart grids alike are made in China. A strategic threat analysis would suggest that this is a bad scenario. In the short to medium term, it is a safe bet to assume that viruses will enter the smart grid and propagate. Our best hope is that we will reach a point where we consider security seriously before one of these threats manifests itself in the form of serious damage.
The real solution is that we need to pay much more attention to the security aspects of the smart grid: real attention as opposed to lip service. It is not beyond the wit of human beings to build highly secure networks and devices. In the longer term, a solution is likely to be found in an approach that mimics human white cell protectors.
We could develop intelligent agents that roam the net looking for bad pieces of software (malware) to be isolated and destroyed. Effectively, a real time anomaly detection and immune system. That – and the decision to take security seriously – should be capable of securing the smart grid.
Professor Peter Cochrane is an entrepreneur and futurologist this excerpt was taken from a contribution to the Engerati Smart Grid report.