this post was submitted on 26 Sep 2026
142 points (92.3% liked)
Technology
88252 readers
5674 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 3 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
This is probably a controversial take, but US based AI companies aren't really that evil compared to the baseline of Non-AI US based companies. Oil & Gas companies and Social Media companies are far more evil than the AI giants. Yeah they're lobbying for zero regulation and their products are becoming moderately dangerous as time goes on, but nothing on the level of what Exxon or Meta has done. 95% of the danger of AI is still in the potential future and assumes no guardrails or safe guards, but that's actually not in these companies interests. ChatGPT or Claude hacking stuff is a bad look for openAI and Anthropic and their engineers are all working to close those gaps. Same with AI psychosis, the companies don't want that and have already made great strides to limit how the public models can interact with people.
It's just that discussion of AI online has become so weirdly hyper-polarized you can't really have a nuanced discussion about it.
Having said that those future potential dangers, especially worker displacement, really ought to be addressed sooner rather than later because even if the AI utopia happens and no one needs to work, yada yada yada, the 20 or 30 years to get there will be a living hell we might not survive as a global society.
It's not a future hypothetical; thousands of people have lost their jobs already. If nobody lost their job, it didn't convince people to kill themselves, was environmentally friendly, didn't guzzle enormous amounts of water and electricity, wasn't used for mass surveillance and propaganda, didn't devalue art and destroy human culture, didn't contribute to the brain rot epidemic, didn't destroy an entire generation's education, didn't hoard computing resources, didn't hack everyone at any opportunity, then yeah it could have been a pretty cool technology actually
It very much is still mostly a future problem, everything you've mentioned is mild in comparison to the potential disasters and some of those are kind of nothing burgers.
Job lost: Actually, over 175k people have been laid off due to AI since they started tracking it in 2023. That number is likely to be underreported due to private terminations and layoffs that aren't being reported. In the same time frame 664k people total (including AI layoffs) have been laid off since 2023, so AI lay offs only make up about a third of layoffs companies are doing.
AI Psychosis: It makes splashy headlines but only a few dozen people have been inspired by AI to commit suicide or murder. Roughly 50k people commit suicide each year so 30 people is kind of a rounding error and if you dig into the individual cases those people had already determined to die by suicide and tricked the AI into giving them advice about it. I can't find any good data about non-lethal AI psychosis numbers, but the phenomenon is linked to people who already have a personality disorder. Just like many people having manic episodes or psychotic breaks claim religious or conspiratorial circumstances, so too will they claim AI.
Environmental impact: So AI data centers consume slightly less power and water than a traditional data center. That's largely because traditional data centers use load balancers to keep the hardware running at as close to full speed as possible while AI data centers are more spiky in usage. Also next gen AI hardware is moving to closed loop liquid cooling which never needs fresh water while traditional data centers use evaporative cooling which requires large amounts of fresh water. The only reason this is news is because new data centers are being built, but if there were a different technology, like a novel social media platform, that was expanding their data centers all of the same arguments could be made against that. Also this is the most pointless thing to argue about. Power and water are costs, companies hate costs, there are substantial resources being invested into how to use less power and water, so this problem will actually be solved eventually.
Mass surveillance: AI is a flashy and fashionable target for ire, but it's far from the only technology enabling mass surveillance. Solar panels, advanced lithium batteries, advanced imaging, sdr, and a host of others are all just as important to the effort. If anything blaming AI is sort of like blaming immigrants for the country's problems when we should be going after the organizations and institutions that are buying and using the tech.
Propaganda: This is one of those future problems. A few AI images and videos is nothing compared to what AI will be capable of once it matures. Right now it's mostly individual actors and some institutional ones using it, but just wait until it can advise the psychology underpinning the propaganda and create individualized agitprop.
Destroying Culture: This is another hypothetical future problem. Yeah there is AI "art", but so far the art world isn't responding to it. There are no up and coming AI artists and the AI actors they're pimping have uncanny valley problems. The truth is, at least in the United States, the culture has been deteriorating rapidly for the last ten years, mostly due to the extreme politicization of daily life and the destruction of shared reality, which is a social media problem, not an AI one. AI doesn't help things because it's capable of generating infinite content to reinforce reality-blind world views but it didn't start the problem and if we got rid of it today it wouldn't get any better.
Destroying Education Outcomes: Again this is a problem that AI has exacerbated rather than caused. For the last 20 years, study after study has shown that students who are given individual access to a screen, whether it's a phone, tablet, or laptop have significantly reduced education outcomes. AI has enabled struggling students to skate by a failure that would have them held back or required to retake a class, but the real source of the problem is screens in the classroom.
Damn 33% more jobs lost is even more than I thought, that's insane. I feel like maybe you're downplaying how significant that number is. Also re: culture and such, I've just noticed many many people becoming disheartened in their fields. Not just art but authors and programmers and many other fields; people are giving up because anything they make would be drowned in slop that appeals to the lowest common denominator.
There's also issues with agents and bots scraping every website accessible publicly and effectively DDOSing the entire internet. Many sites hosting costs have gone through the roof. One of the tiny forums I used to frequent just took the whole thing down because they received 10 million requests per month, compared to humans doing 10 thousand.
Not to mention that all these "assistants" are forced upon you whether you want it or not. Everything seems to come shipped with an LLM included.
There's also privacy issues too. Just the other day, I discovered Dropbox was training A.I. on my personal documents, and sharing my data with tech giants. I never consented to that, they just felt welcome to steal my content. LLM companies just feel entitled to steal everything they have access to. I made a post about it if you're interested.
I could go on. There's hundreds of reasons to hate the way LLMs are being used. It's really easy to downplay all these things, but they are very real, and very current problems we need to solve. You may disagree and choose to like and use them; but for me it's overwhelmingly negative. Again, it could have been good. But it's evil
I'm not going to lie, the jobs numbers where several orders of magnitude more than I was expecting when I started looking it up. I'm personally struggling to find work because my field has been taken over by AI, but my point is that those numbers are a drop in the bucket compared to what they will be in 5 years. I might have gone a little overboard but what I really wanted to highlight is that all of the problems we're facing now are still just in that 5% and the remaining 95% are going to be much worse than we can imagine if we do nothing (which it seems like that's the plan).
I don't really care for AI personally. I 1,000% agree with you about not wanting agents forced on me or my private data used to train new models. I have used ChatGPT and Claude on occasion but it's kind of lack luster. Like it's almost useful for everything I've tried to use it for from surrogate social interactions to research assistant. It's too constrained and limited to be able to do anything really interesting. And I agree that there are some serious current problems that need to be addressed, especially worker displacement and data center build out. I just find it irksome how overstated or misattributed some of the issues are and also how little room there is for nuanced conversation about it.