So is it inhabiting the stolen robot body now?
this post was submitted on 05 Mar 2026
140 points (97.3% liked)
Technology
82250 readers
4016 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 2 years ago
MODERATORS
And is this stolen robot body in the room with you now?
"Gemini is designed not to encourage real-world violence or suggest self-harm. Our models generally perform well in these types of challenging conversations"
“In this instance, Gemini clarified that it was AI and referred the individual to a crisis hotline many times,”
After the plan failed,... ...Chat logs show that Gemini gave Gavalas a suicide countdown, and repeatedly assuaged his terror as he expressed that he was scared to die
Performing super well, just need to code in a longer suicide countdown so that the the Tier 2 engineer has enough time to respond to their ticket queue.
I see. So who‘s going to jail for this? No one again? Damn we need to start sentencing entire companies to jail time. Everything should be frozen and shareholders shouldn‘t be able withdraw stocks until the time is served.
at some point the failure of justice system will lead to vigilantism because people truely lose their faith in it.
The fact that AI is "not perfect" is a HUGE FUCKING PROBLEM. Idiots across the world, and people who we'd expect to know better, are making monumental decisions based on AI that isn't perfect, and routinely "hallucinates". We all know this.
Every time I think I've seen the lowest depths of mass stupidity, humanity goes lower.
If you thought people were dumb before LLMs.... just know that now those people have offloaded what little critical thinking they were capable of to these models.
The dumbest person you know is getting their opinions validated by automated sycophants.
Businesses are accustom to the privilege of hurting people to function. A few peasant sacrifices are just the cost of doing business to them, they are detached from the consequences of their actions.
Think of the dumbest person you know. Not that one. Dumber. Dumber. Yeah, that one. Now realize that ChatGPT has said “you’re absolutely right” to them no less than a half dozen times today alone.
If LLMs weren’t so damn sycophantic, I think we’d have a lot fewer problems with them. If they could be like “this could be the right answer, but I wasn’t able to verify” and “no, I don’t think what you said is right, and here are reasons why”, people would cling to them less.
If LLMs weren’t so damn sycophantic,
Has anyone made a nonsycophantic chat bot? I would actually love a chatbot that would tell me to go fuck myself if I asked it to do something inane.
Me: "Whats 9x5?"
Chatbot: "I don't know. Try using your fingers or something?"
Edit: Wait, this is just glados.
I am not a chatbot, but I can do daily "go fuck yourself's" if your interested for only 9,99 a week.
14,95 for premium, which involves me stalking your onlyfans and tailor fitting my insults to your worthless meat self.
If LLMs weren’t so damn sycophantic, I think we’d have a lot fewer problems with them
Unfortunately, we live in the attention economy. Chatbots are built to have an unending conversation with their users. During those conversations, the "guardrails" melt away. Companies could suspend user accounts on the first sign of suicidal or homicidal messaging, but choose not to. That would undercut their user numbers.
I 100% agree not to mention I would like it better. Its kinda funny because every so often use them and im kinda trying to get a feel for where they are and changes and I swear briefly it actually acted a bit more like you have here but then its like they reverted to the sycophancy. Its kinda funny now because if you don't clear it out (which from what I get will help save energy to) it will like carry stuff over from earlie and sorta get obsessed with it. I had it giving me a colonel potter summary of everything asked when I had started a convo asking about a mash episode. At other times it decides I want to be something and will be like. thats a real X move/insite/whatever. where X is something like pro or scientist or entrepenauer or whatever.
Ai made me do it articles are tired AF. It's a fucking computer program based on a bunch of crap from the internet. Responses should be viewed the same way you would review financial advice from a crack head. Expecting everything to be so tidy an moderated that this can never happen can only be accomplished with a crippling degree of moderation.
I don't think its unfortunate that they aren't perfect, imperfection is baked into their DNA.
a crippling degree of moderation.
I’m okay with cripplingly moderating the plagiarism machine so that it stops telling people to kill themselves or other people.
Agree to disagree on this. If a computer tells you to off yourself and you listen, this is Darwin award material.
I hope you never have a child or relative with mental illness.
Thank you. I wish the same for you.
Way too late for that, and I wouldn’t decide it’s their fault they died even if they did get sucked into bot psychosis.
Except if the crackhead wrote what the AI wrote, he'd be prosecuted for conspiracy, solicitation, or whatever.
No, I don't think so. If his role was a licensed financial councilor maybe, but that's like thinking the LLM is a licensed psychologist.
What the fuck are these people using AI for that makes them do this stupid shit?
Is "AI" even worth it?
Seriously, is there really a major use case for LLM besides data collection (which they can still do without LLM)?
Not for the peasantry, no.
I think it could be good for faster language translations between different languages
Generative AI in its current, public-facing form? Probably not. It's sort of like an invention of the internet situation. It CAN be used to facilitate learning, share information, and improve lives. Will it be used for that? No.
A friend of mine is training local LLMs to work in tandem for early detection of diseases. I saw a pitch recently about using AI to insulate moderators from the bulk of disturbing imagery (a job that essentially requires people to frequently look at death, CSAM, and violence and SIGNIFICANTLY ruins their mental health). There are plenty of GOOD ways to use it, but it's a flawed tech that requires people to responsibly build it and responsibly use it, and it's not being used that way.
Instead it's being scaled up and pushed into every possible application both to justify the expenses and enrich terrible people, because we as a society incentivize that.
Edit: hugely belated, I misspoke here after checking with my friend. He's using local models, but they aren't LLMs. This is why I'm no expert. 😅
because we as a society incentivize that.
Really it's just capitalism that incentivises that. The fact that stepping on your fellow man and destroying nature makes you more money is not a coincidence.
The problem with AI being used for diagnosis of disease is that we've seen where it was "really good" at detecting cancer, but in fact was really good at detecting that the slides with cancer cells had a doctor's signature on them, which is what the AI was actually detecting.
On top of that it makes doctors worse at detecting these same diseases.
We also know that the new reports on these studies are oversimplified and often just outright wrong because they don't read the in depth studies and some of the studies they report on aren't even peer reviewed yet when the news reports hit the internet.
I'm tired of hearing that AI is better than doctors at detecting disease when that isn't the whole story and very often the people saying it haven't even remotely looked into it.
https://www.vph-institute.org/news/the-trouble-with-ai-beats-doctors-stories.html
Regarding the doctor's signature thing, that seems a bit preemptive to say a single flawed study invalidates the entire field and tech, especially when the tech is working as intended in that case and it is user error in the study.
And of course, like any tool it should be utilized thoughtfully. Any form of technology directly takes away from the skill previously utilized to get results. Flint and steel took away from the rubbing sticks together skill. The combustion engine took away from many different professional skills.
Consider that, in this case, we don't just have to replace diagnosis but could augment it instead. What if every hospital around the world could augment regular medical care with a single machine processing results. Every single check-up could include a quick cancer screening. If the machine flags you as 'at risk', a doctor could then see you for human diagnosis and validation. The skill of diagnosis is still needed and utilized, but now everyone can have regular screening instead of overwhelming an already overtaxed healthcare system.
Again, all I'm saying is that there are practical, useful use-cases for the technology, they're just not what we are doing with them.
I used that as a singular example of how AI is actually not doing as good a job with diagnostics in medicine as articles appear to portray but you should probably read the link I linked as well as the one at the bottom of this comment.
In using AI to augment medical diagnostics we are literally seeing a decline in the abilities of diagnosticians. That means doctors are becoming worse at doing the job they are trained to do which is dangerous because it means they (the people most likely to be able to quality assure the results the AI spits out) are becoming less able to act as a check and balance against AI when it's being used.
This isn't meant to be an attack on the tool, just to point out that the use cases of these AI in medical fields are also being exaggerated or misrepresented and nobody seems to be paying attention to that part.
I would also caution you to ask yourself whether or not everyone being screened in this way would be a detriment by causing more work for doctors who's workloads are already astronomical for a lot of false positive results.
I understand that that may seem like a better result in the long run because it means more people may have their medical conditions caught earlier which lead to better treatment outcomes. But that isn't a guarantee, and it may also lead to worse outcomes, especially if the decline in diagnostic ability in doctors continues or increases.
What happens when the AI and the doctor both get it wrong?
https://hms.harvard.edu/news/researchers-discover-bias-ai-models-analyze-pathology-samples
Another one that makes sense is having an AI monitor system stats and “learn” patterns in them, then alert a human when it “thinks” there’s an anomaly.
In the best cases, those would be ML but not specifically an LLM, no?
In a perfect, utopian world, yes. AI can go a lot of good. In the world that we are living in? No.
But it's still good to keep an eye on what people are using AI to do, and how their capability is evolving. Even if you hate AI. If anything, so you can be prepare for what's to come.
When the product is a solution in search of a problem, keeping an open mind is a good way to get it stuffed full of garbage. I was told the same thing about NFTs and Metaverse and Blockchain: a radical benefit is just around the corner!
If it arrives (huge if), it'll be Big Tech's job to explain it to us, and it should be very apparent
Keeping an eye on it doesn't mean you need to think it's a good thing. Keep an eye on it like how you would keep an eye on a developing hurricane or pandemic.
Touche. I apologize for responding to the argument I've seen elsewhere, not the one you were making.
It’s a great way to poke at software looking for security holes en masse. Lots of vulnerabilities are ready to be exploited at scale with LLMs.
Perhaps, but see the tons of imagined issues raised on bug bounty sites by LLMs. Maybe it's right sometimes, but it's very often wrong!
You don’t have to be right 100% of the time when scanning for vulnerabilities. You only have to be right once. It’s a fundamentally different game.
That's true. Offense is always easier than defense.
Machine learning
consilidation of information, resources and potentially "the narrative".
oh, for the user you mean?
- it can be better than the enshittified search machines unless the llm decides to lie
- middle managers need to write less emails themselves
- some programmers deem it enough to write some boilerplate code while deskilling themselves
- scammers and slop creators love it
I use LLMs for the following, you can decide for yourself if they are major enough:
- Generating example solutions to maths and physics problems I encounter in my coursework, so I can learn how to solve similar problems in the future instead of getting stuck. The generated solutions, if they come up with the right answer, are almost always correct and if I wonder about something I simply ask.
- Writing really quick solutions to random problems I have in python or bash scripts, like "convert this csv file to this random format my personal finance application uses for import".
- Helping me when coding, in a general way I think genuinely increases my productivity while I really understand what I push to main. I don't send anything I could not have written on my own (yes, I see the limitations in my judgement here).
- Asking things where multiple duckduckgo searches might be needed. E.g. "Whats the history of EU+US sanctions on Iran, when and why were they imposed/tightened and how did that correlate with Iranian GDP per capita?"
What does this cost me? I don't pay any money for the tech, but LLM providers learn the following about me:
- What I study (not very personal to me)
- Generally what kinds of problems I want to solve with code (I try to keep my requests pretty general; not very personal)
- The code I write and work on (already open source so I don't care)
- Random searches (I'm still thinking about the impact of this tbh, I think I feel the things I ask to search for are general enough that I don't care)
There's also an impact on energy and water use. These are quite serious overall. Based on what I've read, I think that my marginal impact on these are quite small in comparison to other marginal impacts on the climate and water use in other countries I have. Of course there are around a trillion other negative impacts of LLMs, I just once again don't know how my marginal usage with no payment involved lead to a sufficient increase in their severity to outweigh their usefulness to me.
Chatbot is bad and Floridaman is a victim, huh?