The rise of ChatGPT and Midjourney: Are these AI tools more helpful or harmful in society?
Ever since it was coined in the 1950s, artificial intelligence (AI) has become a term that the world seemingly depends on for a lot of things. But it was only recently that it took the world by storm, thanks to the rise of generative AI technology that includes the likes of Midjourney, DALL-E 2, and perhaps the most popular one of all, ChatGPT.
From the word itself, generative AI is made to generate—from photorealistic images to even an editorial piece for a prominent British daily newspaper—through being fed with keywords or "prompts." In simple terms, prompting is how humans talk to this otherwise unseen entity so that they can understand what users want them to do.
This has led individuals who have never even touched a paintbrush before to produce impressive-looking artworks that rival that of the ones crafted through years of practice and skill.
Despite this, it has generated much controversy over recent months, and the question now stands: Does AI offer more help or more harm in this technology-driven society?
New ways to create
Ever since AI has seemingly perfected the ability to generate images and text, people have been using it to satisfy their ingenuity and imagination.
Some have also used generative AI as an added medium to channel their creativity, like technology designer Christopher David who created original Pinoy dishes—complete with recipe names and ingredients list—with help from AI.
To produce the images, David used Midjourney, one of many machine learning-based image generators that have emerged as of late. The way it works is that you must enter short descriptions that are well-written and clear in the chatbox provided. After that, the tool then creates the image based on the instruction.
The same goes for the more prominent ChatGPT, which focuses on text-based outputs such as writing essays, giving relationship advice, answering questions, making customized resumes, and even composing a song.
These types of generative AI models are trained on vast amounts of information from the internet such as websites, books, news articles, existing image samples, including art from various sources, and more.
Sharing his honest thoughts with L!fe in using AI, David couldn't be more excited about the new possibilities regarding the technology. AI has already become an important part of his life, perhaps even a reliable companion.
"A lot of what AI can do now, I never even thought was possible," he said. "I'm excited [about] how much it can even grow, ano pa 'yung mga pwede maging applications, and how it can affect and improve our daily lives."
For digital animator Janina Malinis, however, AI is more of a threat than an aid to artists. Malinis is in disbelief that these technologies are now regularly used by people who aren't even in the discipline of art.
"Part of you is cautious, part of you is guarded na, 'Hala, anong capabilities nito, anong effect nito sa profession mo,'" she said.
Real or artificial?
No matter how impressive it has become, artificial intelligence is still... artificial, but the outputs it has produced have become so realistic that it has managed to fool the world not once, but several instances now.
A notable example of this would be the shot of Pope Francis wearing a gigantic white puffer coat. The picture, created through Midjourney, first surfaced on Reddit before it eventually went viral on Twitter. Many were fooled into thinking it was real.
While the incident was not something to seriously fuss about, AI images eventually became a cause for misinformation that could lead to social unrest.
Case in point, the supposed "death" of a Canadian actor because of cosmetic surgeries to look like BTS member Jimin. Identified as Saint Von Colucci, the figure first emerged from a report by Daily Mail, and it was not until the story was carried by various news sites that his pictures turned out to be a hoax, with one yielding a 75% result on an AI art detector.
Almost every major Korean media outlet ran the story of the "Canadian actor" who allegedly died after getting plastic surgery to look like Jimin. Coverage is still continuing.
— Raphael Rashid (@koryodynasty) April 25, 2023
All are quoting the Daily Mail without any fact-checking. Still awaiting proof this story is real. pic.twitter.com/zVcJabXS2J
Some have also used the tool in other unethical ways, such as students using AI to finish their assignments instead of doing it themselves, with teachers completely left in the dark about it.
It's instances like these that have spurred David to still adopt a cautious approach to using AI despite its advantages.
"With the enormous amount of possibilities on how we can use AI, someone else could use this for terrible things. Nandoon rin 'yun. They could really use it to do malicious things if they wanted to," he warned.
One of the major problems posed by the technology is misinformation. Since tools such as ChatGPT are trained with data that may be outdated, it can give erroneous infomation to its users.
"Even if mali 'yung sinasabi niya, minsan sinasabi niya na tama. 'Yun 'yung problem kasi 'yung iba, gagawa ng essay, ng article, gagamitin nila 'yung ChatGPT pero hindi nila alam na hindi pala accurate 'yung data," David explained.
Adding fuel to the fire is how some people lack comprehension skills to identify which ones are real and which ones are AI-generated, "Ang problema sa Philippines, and daming [mababa ang] reading comprehension, so even if nakalagay na na gawa gawa lang ng AI ['yung post ko about Filipino dishes], 'yung iba akala nila totoo talaga."
"If someone wanted to, they can use AI to really spread misinformation kasi pwede kang gumawa ng picture na mukhang makatotohanan," David said.
Malinis carries with her the same fears, stressing how it's "so easy nowadays to manipulate content, video, photo, and whatever it may be."
Replaced by technology
Another major controversy is how AI allegedly plagiarizes the works of other artists, brought about by how it uses their works to generate art that closely copies those images, which some say is plagiarism.
This has led Malinis to fear that some artists have begun to get exploited, "There are artists nowadays na very distinct 'yung style nila. There's the argument that all art is inspired or derivative, so you can't own a style. While I agree, sometimes kasi very blatant na. May iba na kita mo pa 'yung signature ng artist doon sa AI, nakahalo pa siya."
"Obviously in those cases, it's definitely ripped off and unfair talaga siya doon sa na-plagiarize na artist," she said.
But according to David, the assumption that the technology merely combines pre-existing images to form an output is a misconception.
"AI is based on training. 'Yung mga images na na-train sa kanila, titignan lang ng AI or irerecognize nila ang mga pattern—'yung mga style, brushtrokes, or pixels. Hindi niya sine-save 'yung mismong picture, ang iniistore niya lang ay 'yung mismong training data, kung ano ang narerecognize niyang pattern," he explained.
And now that the lines have become blurry with what is made by AI and what is made by a person, it's only a matter of time before society is faced with unemployment.
"Dati, ang sinasabi ng mga tao, 'yung hindi marereplace ng AI ay 'yung mga creatives. Pero ngayon, dahil sa advancements ng AI, pati sila nanganganib na rin. With ChatGPT, you can just create an article or write an entire book in just a few seconds," David feared.
But Malinis believes that AI still has a long way to go before it could so much as take over her line of work.
"Definitely, I don't think it can replace artists anytime soon. In terms of coming up with compositions or concepts, malayo pa talaga si AI. Mapapansin mo talaga na wala 'yung human touch doon sa mga results," she said.
Regardless, Malinis called attention to the foreboding sense of "unfairness" towards artists as the experience and skill they have have built for a long time had been disrupted by these new AI tools, which are now being preferred by clients as they are faster and more readily available.
Moving forward
Whether we like it or not, AI is here to stay and will continue to make its presence known, but do these types of technology serve for the betterment of society or will they only cause chaos in the end?
"Kung ang i-coconsider lang natin ay 'yung society in terms of kung paano magpro-porgress as a civilization, talagang sobrang laki ng benefits. Wala pa tayo sa kalingkingan ng mga possibilities with AI. But if you look at the individual people within the society, 'yung mga possible na mawalan ng jobs, nandoon yung sobrang laking possibilities for harm din," David said.
In itself, AI technologies bring about no harm to others. The problem ultimately lies with the person who uses these products with ill intentions.
"The fear isn't in the technology itself. It's more of how it would be exploited by people in your circles or even outside the industry, like clients who would take advantage of it instead of making use of local talent or even graphic designers with experience," Malinis said.
"With every tool comes people who use it in the wrong ways and people who use it the right ways. With the tool itself, I don't think it's harmful. I think it's definitely helpful. It's just that it's very easy to use the tool in exploitative ways," she added.
Since AI was only made to improve our lives, it's important that people use it for that purpose only, as David points out, "I don't think we should be scared of AI, but I think we should learn how to actually use it to our advantage. I can use the time na mase-save ko to actually do more things and experiment kung ano pa ang pwede kong gawin."
Malinis urged that people have the responsibility to educate themselves on the right way to use AI tools.
"If you plan on exploring or discovering AI, just educate yourself, make sure that you're using the tools responsibly and definitely not use it for anything with evil intent. Use it with a moral compass. Mahirap na nga 'yung landscape ngayon sa digital misinformation, dadagdagan pa natin," she said.
"You don't use it in any professional capacity without making sure that walang na-harm na artist in the process, like researching about the engines, anong ginamit na library ng engine na 'yun, or did they ask permission or copyright from the artists," she added.
Still, it's not enough that the burden falls on the shoulders of individual users alone. According to Malinis, tech companies and the government should actively map out how AI will change the society.
According to Malinis, ""There has to be regulations. I think there has to be a certain body that carefully notes kung paano nabui-build 'yung engines, paano nila nakukuha 'yung work na naririp-off nila from libraries."
David similarly stressed, "They have to come up with new laws, new regulations, para ma-minimize 'yung risk. It's up to the government and big tech companies on how can they make AI more accessible and how they can make it more safe. Optimistically, ideally, we could use it for the benefit of society."
