Freaky Friday Files: The Horrors of AI
- Ash
- 8 hours ago
- 5 min read
Disclaimer: Information in this blog post may be distressing to some readers. Reader discretion advised.
Hi friends! The other week, Luke and I were watching an episode of Last Week Tonight with John Oliver, and the subject of that week was about AI chatbots, and how f*cked up they are - literally telling people who are s*icidal how they can k*ll themselves, how to make explosives, and how easy it is to break their safeguards, and all of this has stuck with me ever since. Even the creators of these AI tools (usually straight white men) didn't seem to have any empathy for people. My point here is that it's absolutely f*cked up what they are doing, and more of us should be aware, so I was unfortunately inspired to write this blog post about the horrors of AI.
Related Blog Posts:
Why AI Is Problematic?! (4 min read)
The Dead Internet Theory (4 min read)
The Consequences of AI in Everyday Life from a Sociological Lens (15 min read)
Deepfakes: Dangers, Deception, and Digital Disarray (9 min read)
PIN IT
The Horrors of AI (Artificial Intelligence)
I'm sure most of you are familiar with AI, or it's come into your life in some way or the other - it's become SO ubiquitous in everyday life that I think we're really going to have more problems before regulations start to come in. It's becoming harder and harder to get away from, and I get it. Sometimes, it's easy to turn to AI tools in late-stage capitalism, but I just want to say that the men who are creating OpenAI and chatbots and things like that are pure evil, and I am not someone who tends to use the word evil to describe people. I feel like it's worse than it was when social media platforms exploded into our lives. And I like social media, I'm someone who's had a positive experience with social media. I don't doomscroll, but I don't know whether that's because I have good critical-thinking skills or because I'm just not someone who gets addicted to things easily. Anyway, we're going off topic here, but my point about social media and AI is that both these tools tend to be made by men who don't give a sh*t about us as humans.
So let's take a look at some of the horrors of AI.
A person called Matthew Livelsberger (active army soldier who died in this incident), used ChatGPT to help plot his attack on a Tesla Cybertruck outside the Trump International Hotel Las Vegas on the 1st of January, 2025. Livelsberger asked ChatGPT what things would set off certain explosives and general questions about explosives. Livelsberger sh*t himself before the explosives went off.
Encouraging people, mostly teenagers (who are already at risk), to k*ll themselves and giving them detailed plans on how to do so.
The spread of misinformation in general - although does the AI know it's spreading misinformation, or would this just be disinformation? Both are bad. Deakin University has an awesome article explaining this. I HIGHLY recommend you read it "here". Basically, if something is too good to be true, it probably is. Make sure you check your sources, people!!!!!!
Academic students are using AI to do their work for them. This is really, really bad, and I don't get why some universities are allowing this!! Stop it right now! Yes, I am angry.
Exhibiting both political and racial biases - this is not good. It's also interesting where the AI models lean. A study by Stanford University suggested that certain LLMs were left-leaning, which I wonder if it's to do with the researchers themselves, often academics tend to be left-leaning, and they even suggested that maybe the AI models are just reinforcing their confirmation bias. Peters (2002) also says, "Some artificial intelligence (AI) systems can display algorithmic bias, i.e. they may produce outputs that unfairly discriminate against people based on their social identity. Much research on this topic focuses on algorithmic bias that disadvantages people based on their gender or racial identity. The related ethical problems are significant and well known." In a perfect world, AI would only give out neutral opinions. I also want to know if a right-leaning person typed in something into AI, would they also experience confirmation bias? I also think personally that if AIs are racist, then society is going backwards, not forwards, and that's awful.
A robot in South Korea crushed a man to d*ath when the robot mistook the man for a box (how awful).
The National Eating Disorder Association (NEDA) had to suspend its own chatbot when it gave out incorrect medical advice (ew). The chatbot gave harmful weight-loss advice to people who were already experiencing EDs. I think this was in 2023.
Obviously, the people who are making fake p*rn (and even CP) with AI are absolutely f*cked up, and I don't wanna look into that more, so that's all I'm going to say about that one, but it is a problem, especially when it's happening to literal children and teenagers! GRRR I am still grumpy.
In 2024, Air Canada's chatbot gave wrongful advice to a customer securing a bereavement fare. What's even worse about this one is that Air Canada didn't even take accountability and distanced itself from their own chatbot. It went to a small claims court, and I believe the customer won.
A chatbot in New York called My City encouraged business owners to perform illegal activities. For example, paying people less than minimum wage and stealing some of their profits from tips. It really reinforces the rich get richer, and the poor get poorer doesn't it? I do wonder what Marx would think of all this if he were alive today.
In 2023, the magazine Sports Illustrated was using AI to write content.
AI uses so much water in general.
The issue with voice cloning - the AI scams are so disgusting. There are scams out there that someone will call or text you claiming to be a loved one and that they're in danger or need money, and lots of people fall for these scams. I think the best thing to do in these situations is to hang up and call directly to that person to see if it's legit. I think everything is a scam, and although I am definitely not immune.
And that's not even everything that's gone wrong. We can't forget about the blatant plagiarism AI is doing.
It's absolutely terrifying what is happening with the rise of AI. I want to end this post with someone's comment on Reddit, I saw "The real horror starts when the owners of these AI tell everyone (through their AI) what to do, how to act, and what to think. And even a small percentage of the population would create a small army of brain-dead, blind followers." from Reddit user old_man_khan.
Stay safe. Educate yourself. Tell your friends.
Ash xx
References
Campbell, L. O., Babb, K., Lambie, G. W., & Hayes, B. G. (2025). An examination of generative AI response to suicide inquiries: content analysis. JMIR Mental Health, 12, e73623.
Peters, U. (2022). Algorithmic political bias in artificial intelligence systems. Philosophy & Technology, 35(2), 25.
Rozado, D. (2023). The political biases of ChatGPT. Social Sciences, 12(3), 148.
Suchman, L. (2023). The uncontroversial ‘thingness’ of AI. Big Data & Society, 10(2), 20539517231206794.
Winecoff, A., & Klyman, K. (2026, April). From Symptoms to Systems: An Expert-Guided Approach to Understanding Risks of Generative AI for Eating Disorders. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (pp. 1-25).
https://www.ohchr.org/en/stories/2024/07/racism-and-ai-bias-past-leads-bias-future
https://www.naps.edu.au/blog/artificial-intelligence-ai-the-bad
SAVE TO PINTEREST
Comments