The growth in Artificial Intelligence has skyrocketed beyond our very imagination in the past few years. We are beginning to develop models that can perform the tasks of humans and even beat our productivity. Not one, not two but several of those AI-powered systems are beginning to raise worries about AI overtaking humans due to its near-perfect results. This explains why there is a pool of artificial intelligence certification courses to get beginners started in this field.
Google’s AlphaGo project is a simple example of artificial intelligence online. Alpha Go went on to beat the world’s best Go player, Lee Sedol 4 – 1 in March 2016. The ImageNet challenge is another instance of how Artificial Intelligence has achieved human-level possibilities. The best result in 2017 had an error rate of 2.25% which was lower than that of humans, 5% before the competition was canceled.
AI has now galloped from solving ‘mundane’ problems to solving real human problems.
The GPT-2 Breakthrough
The discussion about the advancement of Artificial Intelligence would not be complete without mentioning GPT-2. Folks without a much technological background may as well refer to this technology as blatant magic. GPT-2 is capable of performing state-of-the-art computations in textual data of any form. For instance, given a starting text, the GPT-2 algorithm is able to predict the following words as though it were a human talking. The algorithm was trained on 8 million web pages and its deep learning network contains over one billion neural connections. This network would be more like a mimic of the human brain when used to its full potential. The applications of this would only be limited by our imagination. We would have machines that can write product reviews, write book stories, movie plots, etc. In real-time, programs will be able to identify characters in images or videos, translate texts, answer questions, summarize texts, and even play strategy games such as StarCraft.
Self-driving cars is another mind-blowing technology of Artificial Intelligence. Unbelievably, cars are now making decisions autonomously when navigating the road. Research has shown that these cars are even safer than humans driving. It means beyond the comfort of drivers, autonomous cars are able to avoid crashes, guaranteeing the safety of human life and property. The technology is still in development but has made tremendous progress.
Self-driving cars are not limited to recognizing road networks but other key indicators such as road signs and traffic lights. Once this technology is fully developed, Elon Musk, a pioneer of the autonomous car industry, says that the governmental road regulators would be forced to ban manually driven cars.
But with these advancements, a machine not being conscious will continue to be a bottleneck for Artificial Intelligence. In reality, it is one of the biggest problems of self-driving cars. It is classified as Artificial General Intelligence and it is still a work in progress. There are loads of researches that attempt to distill the latest discovery in this field. A machine that is self-conscious would be able to process billions of data in real-time and perform a wide range of decisions. These are the applications of Artificial Intelligence that can take our jobs away from us. So the question is, will AI ever get to this level.
Will AI attain self-awareness
There are several schools of thought in answering this question. A couple of computer scientists think consciousness in computers will be a stage we attain sometime in the near future. They believe consciousness involves the ability of a computer to receive fresh information, perform some cognitive processing and carry out a commensurate action in response. And with the massive data being generated on a daily basis, this possibility is not so distant. Like humans, machines would be able to store large data in libraries, carry out computations and make complex decisions in split seconds.
On the other hand, other computer scientists are of the opinion that the concept of consciousness cannot be artificially made. Humans have a special entity that makes them creative. It is called the mind. The mind allows humans to think outside the box which causes that sense of freedom that births ingenuity and innovation. Machines, on the other hand, would always only act based on the data it is fed into. There is no such thing as “thinking outside the box”.
Consciousness in 3 stages
To better grasp what consciousness is about, scientists have classified human consciousness into three stages.
- C0: This is the first stage of human consciousness where computations are made in our subconscious without our knowledge. An example would be say during facial recognition. As humans, we do not consciously think about recognizing a celebrity or a friend.
- C1: In this second stage of consciousness, humans are able to make generalized information. In other words, they can receive and filter some quantity of data in a bid to make a specific decision in a specific circumstance.
- C3: This is the third stage of consciousness. It is called self-awareness. Here humans can make decisions in a completely new situation.
You can learn more about AI by signing up for an AI Certification Training Course. In this context, AI has attained C1 and C2 to a large extent. However, there is still a lot of work to be done if we must achieve C3. And in all honesty, it may be hundreds of years from now. In any case, there is no cause to press the panic button.