AI is a known term all over the world by now. Many would agree, in this day and age, that Artificial Intelligence will become smarter then humans in the near future. Elon Musk even warned us all that AI is likely to overtake humans within the next five years. Some may find this idea scary. What will come of us, where do we fit in a world where we’re no longer the most intelligent species. But before we wonder of, imagining ourselves as future slaves of machines or exterminated all together, let us all just first think why we think AI will be smarter than us in the first place.
It is true, AI is exceptionally quick learner. We have seen over the past years that for instance, computer vision getting exponentially better at detecting faces, objects, animals etc. AI relies for a large part on computing power and data. If we consider Moore’s Law, which states that the amount of transistors in an integrated circuit doubles every two years, doubling the computing power roughly every two years with it, AI learns a lot faster than you and me. As for data, more than 90% of all data available has been created over the past two years and is expected to double within the next two years, giving it exponentially more information each and every day.
So AI has a big advantage over us, if we talk about increasing learning capabilities. But there is one major thing that I think we are overlooking when comparing our intelligence to that of artificial intelligence. It is not the same.
I think that we are humanizing artificial intelligence, comparing our abilities to that of the machines. It is true, AI often performs very well in specific tasks. Though when the situations differ, often even slightly, AI has a hard time completing the task correctly. Intelligence is as noted by 52 leading academic researchers in the 1996 wall street journal “a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience. It is not merely book learning, a narrow academic skill, or test-taking smarts.” AI on the other hand is, as mentioned before, very good at performing very specified tasks with specified parameters with specified goals. Therefore, I do not think that AI will become ”smarter than us” so to say. Surely, it will be able to draw certain conclusions quicker, more accurately and have analytical capabilities a BIM master student can only dream of, but it will only be able to do so with certain limits and apply it to reach given goals.
Will it be of no threat to humans in the future than?
That depends of course. How do you specify threat. If we look at our jobs, there are a lot of jobs that are task oriented with specific goals in a given environment. For these jobs, I would argue that in the long run there will be AI systems that will out perform humans and make these jobs extinct. But I also think we humans have special abilities to learn fast, use different kinds of knowledge for different tasks, and think beyond the given parameters and goals and come to unique conclusions and solutions.
If you would look at the threat to our lives. Then I would say that the AI will never be the threat, but the programmers behind the AI might have the strings in hand. These algorithms have a goal they will have to achieve given the limitations, if these limitations are not thought of thoroughly then yes, we might have some scary AI systems among us in the near future…..
References:
https://analyticsindiamag.com/ai-will-overtake-humans-in-five-years-elon-musk/#:~:text=In%20yet%20another%20warning%20against,to%20hell%20in%20five%20years.
https://medium.com/mit-initiative-on-the-digital-economy/how-smart-is-artificial-intelligence-35a384b40a05
https://www.wired.com/story/prepare-artificial-intelligence-produce-less-wizardry/–
https://www.the-next-tech.com/blockchain-technology/how-much-data-is-produced-every-day-2019/
Hey, really interesting topic, it’s been represented several times as a form of entertainment and always seems to focus on the “dangerous” side of machines overtaking us as you mentioned. However, I believe the dangers of AI will not be so obvious and most of them we aren’t capable of imagining at the time. There is the underlying issue that it is a seriously complicated system and almost no one knows how exactly it works since the machines are basically black boxes. That can lead to some problems when the machines begin to be used for functions previously unexplored.
On the other hand, logically, it doesn’t make sense that the machines would have the urge to revolt against humans. The risk of that happening might be worth all the benefits it’s bringing us and will certainly bring more in the future.
As you said at the end as well, it’s essential to make sure the people running these AI algorithms are qualified and realize the power it can bring, which is why heavy regulation will be the way to minimize these risks. Either way, one could argue that these systems are already damaging humans in the form of social media addiction and the mental problems associated with them so the destruction of humans can be seen across multiple lenses.
Hi Henrique, Interesting takes on the topic and I can only agree.There will definitely be loads of problems we could not even imagine now with functions yet to be explored. I would argue though that it is not really the algorithms damaging humans already in the forms of social media addictions and mental problems, but the companies and people behind these systems that make them work this way. The system only operates towards a goal, which it is often very good at, but it is these inconsiderate or incorrectly specified goals that form these problems. I’m sure you would agree.
Thanks for the comment, really made me think more on how these systems actually work and what they are and will be capable of!