How we are Already Committed to the Development of AI
Now that my site has been up for a few weeks and people have had the chance to find it and read some of my work, I would like to get into a more regular release schedule. For the first post of this type, I want to talk about a subject I have seen get a great deal of attention recently, artificial intelligence. Those who have read my book know that artificial intelligence plays an important role, and I find it fascinating to stay up to date with advances that can bring the sort of AI envisioned in my stories closer to reality. Every week we see news stories about AI, sometimes about the advancements being made, sometimes about the fears brought on by those advancements. The fact is that AI is going to play a significant part in the future of humanity as a whole. While it will be some time before we see the artificial life form type of AI science fiction likes to envision, AI is already changing our everyday lives.
The Current State of AI
Virtual assistants, such as Apple’s Siri or Microsoft’s Cortana, can be found on smartphones and numerous other devices that are a regular part of modern life. All of these assistants already rely on rudimentary AI algorithms. Like these virtual assistants, most AI will be little more than advanced tools; designed for specific purposes and aiding people in carrying out specific tasks. This type of AI is of particular interest for its potential in fields such as material engineering or pharmaceuticals, where it can help develop new materials or medicines respectively. Yet, development of AI with capabilities rivaling or even surpassing those of humans is on the horizon.
Since the 1960’s computing power has tended to double every two years. However, this trend has been slowing as we approach the limits of miniaturization. By the mid 2020’s we may reach a point where components are designed on an atomic level and thus cannot be miniaturized further. This is one of the factors driving firms such as IBM and Google to develop quantum computers which have the potential to be millions of times more powerful than conventional computers. While quantum computing is still in its infancy, it offers staggering possibilities for numerous fields, including artificial intelligence. As quantum computing advances it is entirely possible that we may be able to create the very advanced forms of AI found in science fiction.
For those interested in a more detailed explanation of quantum computers and the principles behind them, a good explanation of the fundamentals can be found here: http://www.research.ibm.com/ibm-q/learn/what-is-quantum-computing/
AI of the Future: Dangers and Possibilities
Along with the prospects and possibilities of advanced AI comes the inevitable fear of the AI apocalypse. From books and movies, to scholarly articles and governmental debates, the prospect of murderous AI annihilating or enslaving humanity seems to captivate us. Tech billionaires and scientists repeatedly tell us that AI will destroy us, while activists call for bans on killer robots. The question is whether this is fear mongering or a legitimate concern. Much of this fear seems to stem from our tendency to ascribe human traits to AI. We project a whole host of human motivations and traits to any AI we envision, in science fiction AIs are often given a human form even if there is no practical reason behind it. In most circumstances, there would be no reason for an AI to have such traits or motivations; they would be superfluous or even counterproductive. Perhaps the most pervasive of the motivations is self-preservation.
Many visions of the AI apocalypse are predicated on an AI having an instinct for self-preservation, and “logically” deciding to wipe out humanity because humanity would no doubt destroy it. Even if we assume that no one would program an AI to preserve itself at all cost, we hear fears of an AI learning at an incredible rate and developing such desires on its own. However, this is once again ascribing human traits to the AI. Simply because an AI understands the concept of self-preservation does not mean it will necessarily have the desire for it. Furthermore, should an AI develop a true desire for self-preservation, the assumption that it would naturally decide to wipe out humanity seems to lack a logical basis and instead be rooted in our own fears of the unknown. The truth is we simply do not know what will happen if and when we create an AI that transcends human intelligence, but that gives us reason to be cautious, not outright abandon the possibilities.
Ultimately, it seems that the greatest threat from AI in the near future will come from bad programming, either as a result of negligence or intentional maliciousness. It is the possibility of intentional malicious programming on the part of humans that we should be most concerned about when it comes to AI. We already live in a world where dealing with computer viruses and other forms of malware is a constant battle, and the threats posed are only multiplied when one considers the damage a malicious AI could do. Unfortunately, that threat does not seem to be one that can be avoided. AI is going to be a major influence in the future regardless of any potential dangers it poses, and we must advance our understanding of AI so we are prepared to deal with those dangers.
But often overshadowed by the fears surrounding AI are the benefits it could provide. The advancements to medicine alone could revolutionize the future. With AI it becomes easier to design specific treatments for an individual; to design drugs that target specific problems with minimal side effects. Combined with advances in quantum computing it could allow us to solve problems and run simulations that would be impossible with current technology, expanding our knowledge in ways we may not yet imagine. No matter what, the development of AI is not something that can be stopped, it already exists and it is only going to become more advanced. In a world that has become inextricably tied to technology, AI will play an ever expanding role in our future.
