This post was originally published by Rahul Dev at Medium [AI]
Artificial intelligence becoming more and more common in national defense, it is vital to understand both the advantages and disadvantages of using artificial intelligence in cyberwarfare.
It is also important to know what advantages and disadvantages there are for actual human interaction when discussing cyberwarfare. It is the job of this paper to show you the pros and cons of both of these and the affect they have on fighting wars, both physical and cyber.
With a huge increase in cyber-attacks over the past couple of years, we are seeing more and more companies AND governments turning to artificial intelligence to defend themselves more effectively. So, let’s explore some advantages of using artificial intelligence. First off, many military agencies use artificial intelligence for unmanned undersea and aerial vehicles that can use the weapons on board to attack certain targets (Jenson, 2016). The benefits of these are that they are relatively low- cost and they don’t put any lives at risk (aside from the person being targeted) because a person does not need to be inside it when it is driving or shooting.
Another advantage of artificial intelligence is the fact that these systems have the power to detect and identify threats on the network much faster than an average human. Now with applications, cloud services, and mobile devices (e.g., tablets, mobile phones, Bluetooth devices, and smart watches) there is a much bigger attack surface to defend (George, 2017). But, with these machine learning systems, they can analyze data much faster than humans and are able to detect threats right as they occur. Artificial intelligence can also help in defending our systems as well. AI technologies are getting so sophisticated that now once an attack happens in cyberspace, these systems can instantly start to put up barriers and “cognitive” antivirus systems (Ashford, 2016). This means that these systems can identify, analyze, learn, anticipate and defend cyber security threats on a daily basis (Ashford, 2016). The last advantage of AI is something called “cognitive electronic warfare.”
The traditional way of electronic warfare relied on human interaction and took way to long. Now, An AI system can pull data from aircraft in flight, detect a new threat, and automatically issue appropriate updates to every aircrafts software so they can counter it. The AI system also can update maps of where threats are and automatically reroute both manned and unmanned aircraft around them (Freedberg, 2016).
Moving on, we now get to the disadvantages of using artificial intelligence in cyberwarfare and war in general. First off, there is the risk that one of these systems could malfunction. If this does happen the consequences might be minor, but, there is also the chance that they could cause devastating effects or the destruction of mass amounts of data. For example, an AI system in charge of defending the White House network could malfunction and accidently shut down the entire network leaving the entire White House staff and the president without the ability to get online or communicate with the outside world. Now I know there are systems in place to prevent this, but every system has its weakness.
There is also the fact that it is very possible that these AI systems could end up in the wrong hands which brings a potential for even greater cyber security threats than we have now (viaresource). Another disadvantage of artificial intelligence is that these systems don’t have the power of complex thought. We cannot rely on these machines to make morally just decision on things because they don’t have the power to understand the concept of morality. Lastly, there is the thought that one day these systems might become too intelligent. This could cause humans to become obsolete and no longer needed on this planet.
We now move on to the advantages of human interaction in cyberwarfare. Humans have a distinct ability over AI in the fact that we can have various responses to different threats. We can change a plan if we feel that there is a safer option that causes less damage and is beneficial for all parties involved. Our ability to have complex thoughts gives us an advantage over artificial intelligence because we have the ability to create new things like more secure software, or faster computers, or better defense systems. Humans also don’t work in a singular motion like artificial intelligence does. We have the power of creativity and ingenuity.
This is an advantage because we have a better understanding on planning things when it comes to cyberwarfare and war in general (Jankel, 2015). We can predict moves that enemies are trying to make and we can use trickery to fool them and defend ourselves. Another advantage that humans have is the power to make the right decision. For example, if you need to get somewhere as fast as possible in a car, an AI might drive you 100mph all the way to your destination. But, we have the ability to make proper decisions and follow traffic rules and regulations.
We lastly get to the disadvantages of human interaction in cyberwarfare. To begin, humans are emotional beings. This is a problem because we can easily let our emotions get the best of us and make the wrong decisions. When talking about cyberwarfare this could cause disastrous results in attacking a foreign nation just because you are angry at them. Humans are also rather slow when it comes to decision making. In The cyberworld the slower you are to respond to something the more devastating the effects will be to your network. Lastly, humans cannot be awake and alert 24/7. We need to sleep in order to function properly, we cannot be defending our network all the time with no rest.
Although I do have a lot of confidence in AI for conducting cyberwarfare, I still have some doubts about it. Intelligence systems are amazing at defending networks and providing us with data about who is attacking, what they are attacking with, and the steps we should take in order to defend our networks. What I am a little worried about is when AI starts to attack other entities. My concern is that because these systems lack complex thought, they don’t have the ability to think out the situation and ask the question of is it worth it to attack? Or what will the consequences be when we attack? So, even though artificial intelligence is making life a lot easier for us humans, we have to be careful how much we let these systems control.
This post was originally published by Rahul Dev at Medium [AI]