In today’s society with such improved technology, artificial intelligence (AI) is on the exponential rise. AI is getting smarter and faster than we can comprehend, from simple photo improvements to fully AI-generated videos and websites. Generated products and descriptions are among stores across Amazon, Temu and AliExpress. According to a study by Mark Webster in 2024, Americans lost over 100 million dollars to generated scams. According to the study, the most reported scam was AI fraud that used technologies like deepfakes and voice cloning. A deepfake consists of taking several pictures of a face and using pattern recognition, to create a “mask,” which is placed over an existing video. Voice cloning is a similar process where you take several samples of a voice and then it uses that audio to create a mimic of that voice.
However, both technologies require a lot of imputed data to use, but even with an unlimited amount of data, the AI will still make mistakes, which you can use to determine whether an image or video is generated or not.
“Try to look at the details. These days, it’s harder with photos but still relatively easy with video. But if you look at something specific over the course of a shot and see how it moves, how it changes, what happens to it, you might find things that don’t really make much sense,” Visual Effects artist (VFX) and engineer Wren Weichman said.
Recently, scammers have been using website creation tools like Squarespace, Wix, and GoDaddy. All of these sites incorporate AI tools for making websites faster to make “products” that are completely fabricated by image generators like Midjourney or Stable Diffusion. An example of this is lots of products from Temu. On Temu, it is so easy to find items that are fake brands or designs because, in essence, Temu is what’s called a dropshipping site, which is an outlet for smaller businesses to get their products out and advertised. However, there’s no quality control which is the act of making sure that the product advertised is what you are getting and that it’s of an acceptable quality for retailers, which makes these generated products extremely easy to get tricked by, especially when the images are completely different from what you get.
The differences between generated videos and photos are substantial, when AI first started generating videos, they were like abstract paintings; they had constantly changing details, and like the generated images, the details don’t stay consistent. But now, as video generators get better, it’s getting harder to tell, but it’s still possible.
“In videos sometimes there’s a little bit of a blur, you can look at skin, like ultra smooth or textured skin, if they have six fingers, AI has a lot of trouble with hands, hands will look deformed and they’ll have more or less than five fingers.” CTE teacher Michael Poller said.
For students and educators, AI can be grade-saving or grade-dropping. Apps like Photomath and Gauth have both been using open-source intelligences like OpenAI and TensorFlow to take photos of questions from a variety of subjects and output an answer. However, teachers have caught on to the use of these tools, using a mix of human senses and essay checkers like Grammarly and ZeroGPT.
We as humans have a keen sense if something is off-putting to us in pictures or videos of people or common things that we see day-to-day like someone’s writing. If something gets marked in our mind as strange then it’s an effect called the “Uncanny Valley Effect.” AI Videos often get called uncanny, and there’s a good reason.
A popular example of a widely known generated video is the Coca-Cola ad called “The Holiday Magic is Coming.” The advertisement features shots of Computer-Generated Images (CGI), real, and of course, AI footage. If you watch any part of the ad with the Coca-Cola trucks, it simply doesn’t make sense, the number of tires doesn’t always stay constant; instead of driving over the ground, they glide, and the trucks that are further in the background begin to morph and contort. Coca-Cola hasn’t attempted any more AI-generated ads because of the severe controversy that this ad stirred up. This controversy put a lot of public eyes on AI and has led to increased amounts of disdain.
A common fear of artificial intelligence is its ever-increasing rise to power and its fast gaining of knowledge which could at some point take jobs away from real people. It’s growing at an incredibly fast rate.
“The reason is exponential growth. New advances unlock even newer advances, which unlock even more. It’s also a bit of an arms race currently, so there’s extra demand for getting to the cutting edge right now. This, in turn, creates advancement that seems way faster than we’d expect,” Weichman said.
AI has many, many, many uses, from the people who use it for school or work, or the people who use it to create artwork for an easy cash-grab to the people who think that they can have a personal relationship with an artificial intelligence chatbot. AI is here to stay in our lives and is only going to get more prevalent. AI will get harder to detect even by those with the highest expertise.
“ChatGPT & Magic School AI are the 2 I have experience with. The way I use them is with questions that I may need to come up with from my research and reading of my lessons. ChatGPT is more of a structure of questions. Magic school AI is more for educators, so I use it for teaching lessons or planning activities to help hit the instruction .” CTE teacher Mac Porter said.
AI is getting developed into almost every minute part of our lives, like with the NEO Artificial Intelligence-powered robot from 1X. The NEO can deep clean and manage your home and even hold conversations, help you remodel, or tutor you. The abilities that the NEO has, could put maids, home planners and tutors out of work. Artificial intelligence is also being tested and used in the world of law and legality.
According to Harvard University, artificial intelligence in the world of law is becoming more widespread as it has been documented that there are talks of judges being replaced with AI because it’s much faster than a traditional judge and cannot be bribed.
Getting scammed by a regular person can already be devastating enough, but if AI continues getting smarter the way it has been, then the number of people who get affected by these scams will also begin to rise exponentially. Remaining vigilant could be the difference between an annoying ad and a loss of several thousand dollars.
