Coexisting: Learning to live in a world with robots

The ancient Greeks used the catapult to launch arrows, darts and stones at enemies during war times. The Ancient Egyptians created the lever to assist  in construction. The computer was invented to connect people, provide entertainment and obtain any discoverable knowledge within a second. 

All of these machines, and many others, show the harsh truth that most anything humans can do, machines can do better. Ever since the invention of the computer, artificial intelligence has become a vital part of our everyday lives. 

Through the development of AI, the world of sci-fi movies and books have run with the idea of the inevitable robot uprising. Movies such as “I-Robot,” “Chappie” and “2001: A Space Odyssey” have helped develop the idea that humans will soon live in a world dominated by man-made machines. With those exaggerated fiction stories, people have quickly created a newfound fear that AI is preparing to take over.

So why does humanity fear AI? While humans are in control of the machines we create, the common fear is that due to the human-like intelligence these machines hold, they will soon develop a way to go beyond their limits to outsmart and overpower humans. 

Within the past couple of years, groundbreaking technology has emerged making society fear AI more. The invention of chatbot ChatGPT took the world by storm and has completely changed how education is viewed. 

In Nov. 2022, research laboratory OpenAI launched ChatGPT. The company did not expect the chatbot to flourish in the months following its release. Though ChatGPT is used by the public as a way to discover information across the internet, that was not its intended design. It was originally created to hold a conversation with its user. The common belief that ChatGPT works by conducting a search across the entire internet and filtering out key words is inaccurate. ChatGPT actually has no access to the internet itself and instead uses information that it learns from training data to provide the user with a response. The data provided stems from the internet, books and other sources. This is why ChatGPT cannot always provide the correct information: it does not know everything that the internet knows. While ChatGPT will write poems, emails, code and nearly any other prompt you ask it to, if you are looking for the most reliable information, sticking to a search engine such as Google would be the best choice. 

Another equally popular creation by OpenAI is Dall-E. After receiving a prompt, Dall-E compiles images from across the internet to create the image you asked for to the best of its ability. Similar to ChapGPT, there is an ethical dilemma with Dall-E. When creating the program, OpenAI fed the system pictures from the internet, books and other forms of literature. Since it creates new artwork with already existing art, copyright can be questionable as the art is compiled from various resources, all previously created by another artist. 

So does combining pictures to conjure AI output violate copyright laws? It doesn’t. In a decision made by the U.S Copyright Office, they stated that AI generated images are not protected under current copyright laws because they “are not the product of human authorship.” So while it is completely unethical to use and publish AI generated images, it is not illegal. 

In most cases involving powerful AI, companies do not intend to create machines to be used maliciously, but the public often uses them in dangerous ways. 

Imagine someone using technology to mimic your voice, face, and mannerisms to create a video of a person identical to yourself doing whatever they desire. This technology, known as deepfakes, has existed for several years. The most technologically advanced method of producing deepfakes involves filming a video of a person doing or saying something and superimposing the person you’re trying to mimic onto them. 

This type of technology can be dangerous. A Reddit user used the technology to paste new faces onto pornographic videos, leading to the coining of the term deepfake. Another worrisome use of deepfakes is when creators make videos or audio clips of politicians. In 2020, a video circulated of the Belgium Prime Minister linking COVID-19 to climate change, even though this was AI formed.

It is not new for humans to question the intentions and capabilities of AI. In 1950, after people began to fear the idea of coexisting with an artificial brain, computer scientist and mathematician Alan Turing posed the question “Can machines think?” He followed this question with an experiment where a person unknowingly has a conversation with a machine. If the machine can contribute to the conversation and fool the person into thinking they are speaking with a human, it is able to show human intelligence and thus passes the test. 

Some scientists view the Turing Test as invalid due to the lack of an explicit definition of “intelligence” and the variety that can occur during the test. It has proved to be quite difficult for developers of Artificial Intelligence to create a machine that can successfully pass the Turing Test but is something more developers strive for. 

Through Turing’s thought process and test, the world began thinking about AI in a more theoretical and complex way, they were no longer just machines created to help aid society. 

Yet, as proven time and time again, AI will not rise up and take over the world. Artificial intelligence has simply become machines that are a must have for human aid. While we are capable of completing most anything a computer can, life has become much easier with their assistance. There are many reasons why AI will never replace humans, one of the most prominent being that machines lack human emotions. Without emotions, many of the jobs that humans hold would be pointless. 

AI is always evolving and with new technology such as ChatGPT, we can’t help but wonder what comes next. It’s hard to imagine a company that AI has not drastically changed or affected. 

When trying to think about how much more AI can do, it helps to consider all the things that it still can not do. Yes AI is considered to be an artificial brain, however, these machines are still very different from humans, and a lot of steps need to be taken for them to be. Self-driving cars allow a driver to sit back and relax on the way to their destination, but they still require a person to be there in case something goes wrong. They are not fully autonomous yet. There is certain technology that you can attach to virtual assistants such as Alexa that can make her turn off your lights or television, but you can’t ask her to make your bed.

Yolanda Gil, AAAI President at the University of Southern California, and Bart Selman, AAAI President-Elect at Cornell University designed a 20 year AI Roadmap showing what they believe research will be dedicated to. In this, they predict that AI will help reduce healthcare costs, accelerate scientific discoveries, and help national defense. They believe so much progress will be made because now more than ever people are continuing to build AI machines, and data continues to grow in discoverability.

Scenarios humans create in which AI develops the knowledge to overthrow its creators are disproven by how machines actually work. AI can only function based on the information it is given. It is nearly impossible for a machine to go beyond what it was coded to do. Without human intelligence, artificial intelligence would not exist, and it is completely reliant on humans. So while there will not be a robot uprising that will leave our world in an apocalyptic state, humanity will continue developing cutting-edge new technology that will shape the future before us. AI is ever-changing, and rather than fearing the outcome, we should learn to embrace it because artificial intelligence is not going anywhere.