It’s December 2022. I have been going around all year figuring out the curiosities of life and technology around me for the past few months since I wrote an article or posted tech tweets and stuff. And of course, the gruesome hours of college life as well. And I have made great discoveries during that period, some of which I wish I documented that time.
Let’s be real; hasn’t technology evolved a lot? Well, it is still a long way to go. Nothing is 100% complete unless it’s fully functional and operational. That’s what tech is about, it must be perfect when it serves a genuine purpose. And if not, it misses the mark. And just as tech YouTuber Marques Brownlee said, “Tech is so great when it works……” Somehow intangible tech is fascinating when it becomes a hot topic. That’s why I write this article before I get to write any other article before the year ends.
The subject is Artificial Intelligence (AI). It is a course unit I personally pursued this semester, and it has very wide knowledge areas, some of which I didn’t get to learn more till then with additional coursework of course. AI (as abbreviated) is simply the science involved with the design, implementation and maintenance of systems that communicate with their environments through traditional human sensory means and proceed intelligently without any human intervention. They can mimic human behaviour in terms of reasoning and decision-making.
AI is used in common and newer areas of research; in robotics, expert systems, natural language processing, gaming and automated decision-making. The goals of AI are numerous in number, the top goal being ‘To replicate human intelligence’. Some might say, it solves knowledge-intensive tasks, while others may have a conclusive remark stating, it has an intelligent connection between perception and action. All of the above makes a true summary of Goals in AI. Now, AI is evolving, and more goals keep coming up to simplify lesser intensive tasks from day to day.
As I mentioned earlier, AI is a branch of computer science and also constitutes a tool for users and developers. Let me explain, AI can be used to analyze your Google search results, which goes hand in hand with a subset called Machine Learning (abbreviated as ML). Say you’re searching for the term ‘label’ and your results appear. The general results may appear as the most recommended. Now search using the words ‘what label….’. A list of suggested common searches will appear and luckily enough, you might find one result with your search: What label does Colgate have. It gets to learn that you’re looking for a specific answer with the same term used previously, relating to your intended area of focus — ‘label’ by categorizing and predicting.
Google searches aside, there are many use cases of AI:
- It is present in social media, like on Instagram and TikTok AI considers your likes and the accounts you follow to determine what posts you are shown on your explore tab or Four Your Page (FYP) through a generated and filtered algorithm.
- It is used in email providers, where the email that we use in our day-to-day lives has AI that filters out spam emails sending them to spam or trash folders, letting us see the filtered content only.
- Technology devices like phones, tablets PCs and laptops have camera sensors that are embedded with AI systems. Our favourite devices use facial recognition techniques by using face filters to detect and identify to provide secure access. Apart from personal usage, facial recognition is a widely used Artificial Intelligence application even in high security-related areas in several industries.
And those are just some of the common use cases. And then there became new ones…
I can write a lot of points appreciating what AI can do, however, due to the recent events of the AI trends, I had to research deeply on the areas of interest and why the sudden shift of trend. When I mentioned that AI is sometimes used for lesser intensive tasks, I was exploring some of the new and current tools, apps and websites that have become publicly available and accomplish those goals. One that caught my attention is OpenAI and it is the focus of this piece. Then there was the emergence of AI apps in the app stores.
OpenAI is an Artificial Intelligence research and deployment company founded by Elon Musk on 11th December 2015. Its dominance has been greatly noticed by the public since 2021 for building AI Models. OpenAI is currently known for products like DALL-E and Chat GPT. DALL-E (2) is an AI tool or website that generates images from descriptive text whereas Chat GPT (3) is a prototype artificial intelligence chatbot developed by OpenAI which specializes in dialogue. While the details of pure AI might sound deep and complex to analyze and understand for basic knowledge, OpenAI at least seems to simplify its complexity it. Most companies are relying on AI Tools for research, productivity and various task forces. Developers from any level use Chat GPT to understand and rectify the code error they missed so that deployment can take place before the final submission.
So then, what is the craze about? Why now in 2022? The trends have become fascinating to look at and experiment with them recently around the internet. By the way, did you know that AI is also involved in making music? Neither did I. So, I did my research about AI and how it is applied in music using a software called Verbasizer. It was crazy to have that sort of creation in the beginning but ended up being an adaptive norm in the industry. Of course, this is not a new trend but a recent one of events. The real interesting trends are Chat GPT and DALL-E. Now DALL-E alternatives have been flooding various app stores. Apps like Prisma, Dawn and Lensa AI are not word-descriptive to generate artistic images, but instead, compute and process a ready image from you. These images create cool avatar art that looks like they were actually ‘drawn by humans’. It is impressive to be honest, even most social media influencers and photo genetics are using it to live up to the hype to show their artistic imagery.
AI is cool and all, but it seems like a lot of troubles, worries and legal issues have resurfaced recently, hence the pros and cons. When it comes to Generative AI Tools, it is Big Data that also plays a major role in conveying GIGO (Garbage in Garbage Out) techniques, especially since it is publicly used. You feed in data; you expect information to be presented to you to meet your satisfaction. That satisfaction might not be 100% accurate especially when you use Chat GPT since some information is considered out of date and this source of information is probably uncertain. In short, not every result is 100% guaranteed and that is considered a limitation.
Students have been relying on Chat GPT to do research, check for code errors and summarise some literature which honestly has been ideal and reliable for the coursework we take. And I testify to this, it has been close to two months since I got to play around with DALL-E, OpenAI and Chat GPT. DALL-E image results always differ from what I envisioned that image to be, OpenAI has a lot of limitations in getting the coding answers I want to compare with and Chat GPT on a good day has a lot of Internet traffic when I want to ask a question. The latter is weirdly relatable on campus at a certain time when I want to make sure the content, I am looking for is crucial to my coursework. Suppose the developers find such limitations in coding or personal research, it is not worth relying on it at all.
Unfortunately, most of the human artists who have contributed their skills to the platform do not get full credit or none. Copyrighted art is fed into DALL-E and you will probably notice some of the individual’s art signatures. So, I wonder as a graphic designer, would my work of art be fed into DALL-E one way or another, without my consent? Is this how digital artists feel when their work is robbed of? (I still do not advise you to do NFTs to own your art. Story for another day). So basically, AI stealing art without consent is one problem that needs to be addressed, but what about the fears of AI replacing human labour, like robots taking over?
For your information, this is not the first time AI has become a hot topic. In 2018, researchers have been tirelessly trying to understand the system of AI and what pros and cons it brings, and it is a lot to dig into. That is still the case today. So, should we rely on AI for dependency? Data safety may or may not become a part of it so, only time will tell when major companies and industries get to explore the possibilities of AI.
It may be too early to reliably measure the impact of AI on society — the industry is only just getting started — but preparing ourselves for what it all means and how it will affect daily life, work, and public institutions like health care, education, and law enforcement is perhaps just as important as the research and product development itself.
