Disclaimer: I am asking this for a college class. I need to find an example of AI being used unethically. I figured this would be one of the best places to ask. Maybe this could also serve as a good post to collect examples.
So what have you got?
the IDF using it to carry out airstrikes on Palestinians https://en.wikipedia.org/wiki/AI-assisted_targeting_in_the_Gaza_Strip
Ow. Fuck.
I’m a university prof in a medical science field. We hired a new, tenure-line prof to teach introductory musculoskeletal anatomy to prepare our students for the more rigorous, full systems anatomy that’s taught by a different professor. We learned (too late, after a year) that they used AI to generate the slides they used in lecture and never questioned/evaluated the content. Had an entire cohort of students fail the subsequent anatomy course after that.
But in my mind, what’s worse is that the administration did nothing to correct the prof, and continues to push a pro-AI narrative in order for us to spend less time investing resources in teaching.
Jesus fucking Christ. That’s horrifying.
Not exactly AI being used, rather developed, but Meta’s torrenting 80tb of books and not seeding is egregious
The fact that so much training data is scraped without consent makes a lot of the popular LLMs unethical already in their development, yeah. And that in turn makes using the models unethical.
Flooding the internet with slop.
Mass consumption of copyright works for training, but still considering individuals doing it to be criminals.
Using it at all, really. Given the environmental costs, the social costs, and the fraud it entails, using it at all is pretty much unethical.
My favourite example, though, was the lazy lawyer who used ChatGPT to write a legal brief for him.
Given the environmental costs, the social costs, and the fraud it entails, using it at all is pretty much unethical.
There are loads of examples of AI being used in socially positive ways. AI doesn’t just mean ChatGPT.
Excelled point. I think there are some legitimate uses of AI. Especially in image processing for science related topics.
But for the most part, almost every common use is unethical. Whether it be the energy demands, (and its contributions to climate change), the theft of intellectual property, the spread of misinformation, and so much more. Overall, it’s a huge net negative on society.
I remember hearing about the lawyer one. IIRC chatGPT was citing laws that didn’t even exist. How do you not check what it wrote? You wouldn’t blindly accept predictive word typing with your phones keyboard and autocorrect. So why would you blindly trust a fancier autocorrect?
But for the most part, almost every common use is unethical.
The most common uses of AI are not in the headlines. Your email spam filter is AI.
You are being a little bit pedantic. People talking about “AI” today are talking about “LLMs”, not the older tech that turned out not to actually be “AI”. (Rather like the current stuff isn’t actually “AI”.)
Not supervising your Tesla properly and running over people.
Maybe “favorite example” isn’t the best phrasing for the question, but I get the sentiment and would have to say using AI to create porn of real people as a means of blackmail.
I Used to Teach Students. Now I Catch ChatGPT Cheats
https://thewalrus.ca/i-used-to-teach-students-now-i-catch-chatgpt-cheats
Students using it as a way to avoid actually learning and learning how to think for themselves. Also for being yet another soul-crushing blow for already under-paid, under-respected, under-attack teachers.
I’m a month away from my IT diploma. Even the teachers are feeding us AI slop at this point.
They gave up trying to get the students to stop at the end of first year. Protip: don’t hire a new IT grad, they don’t know anything chatGPT doesn’t know.
I interviewed a candidate recently, and they basically lost all consideration when I asked them a basic sysadmin question and they replied, “That’s kind of one of those basic commands I just ask ChatGPT.”
The basic sysadmin question was: “Name one way on a Linux server to check the free disk space”.
Sadly, I had to continue the interview, but I didn’t even bother writing down any of the candidate’s responses after that. The equivalent would have been asking them “what’s 2+2?” and they break out a calculator. Instant fail.
Someone else commented in another thread.
His sister is in 3rd grade and used AI to answer 'How many seconds in three minutes?"
Goddamn.
I know teachers can’t do this, but they should be allowed to be like: ChatGPT, you get an A. Susie, you will be repeating 3rd grade.
This is why the folks in Silicon Valley don’t let their own kids have tech.
https://www.snopes.com/fact-check/tech-billionaire-parents-limit/
I read that one the other day. Unbelievable that tertiary students aren’t there to learn.
Thank you for asking actual humans instead of an LLM 😊 Here’s my favourite example, and it’s worth digging into more: https://pivot-to-ai.com/2025/03/10/foreign-policy-the-u-k-pivot-to-ai-is-doomed-from-the-start/
It felt like the right way to approach the topic. AI has become so pervasive, I’m not even sure I could search for it without simultaneously using AI.
Is erotic roleplay ‘unethical’? Because we got a lot of services for that.
