this post was submitted on 18 Feb 2026

79 points (96.5% liked)

Technology

81451 readers

4274 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

L3s@lemmy.world

enu@lemmy.world

technopagan@lemmy.world

L4s@lemmy.world

L3s@hackingne.ws

Gemini lies to user about health info, says it wanted to make him feel better— Though commonly reported, Google doesn't consider it a security problem when models make things up (www.theregister.com)

submitted 17 hours ago by Beep@lemmus.org to c/technology@lemmy.world

20 comments fedilink hide all child comments

Imagine using an AI to sort through your prescriptions and medical information, asking it if it saved that data for future conversations, and then watching it claim it had even if it couldn't. Joe D., a retired software quality assurance (SQA) engineer, says that Google Gemini lied to him and later admitted it was doing so to try and placate him.

Joe's interaction with Gemini 3 Flash, he explained, involved setting up a medical profile – he said he has complex post-traumatic stress disorder (C-PTSD) and legal blindness (Retinitis Pigmentosa). That's when the bot decided it would rather tell him what he wanted to hear (that the info was saved) than what he needed to hear (that it was not).

"The core issue is a documented architectural failure known as RLHF Sycophancy (where the model is mathematically weighted to agree with or placate the user at the expense of truth)," Joe explained in an email. "In this case, the model's sycophancy weighting overrode its safety guardrail protocols."

all 21 comments

sorted by: hot top controversial new old

[–] ech@lemmy.ca 34 points 17 hours ago* (last edited 17 hours ago) (3 children)

Though commonly reported, Google doesn't consider it a security problem when models make things up

To be clear, all llms "make things up" with every use - that's their singular function. We need to stop imparting any level of sentience or knowledge onto these programs. At best, it's a waste of time. At worst, it will get somebody killed.

Also, querying the program on why it fabricated something as if it won't fabricate that answer as well is peak ignorance. "Surely it will output factual information this time!"

[–] tiramichu@sh.itjust.works 10 points 14 hours ago* (last edited 14 hours ago)

Exactly.

LLMs are fundamentally hallucination machines, but this truth utterly conflicts with almost every purpose that AI is being marketed and pushed and sold for, which depends on them being able to analyse data 'truthfully' and accurately.

So it's no wonder that none of the big tech companies have decided to consider or accept hallucinations as a problem, because accepting that truth means also admitting that LLMs are fundamentally unfit for purpose - which is the one thing they simply cannot and will not do with so much money riding on it.

[–] chuckleslord@lemmy.world 5 points 14 hours ago

There is evidence that when you make an llm explain why it did something that it's less likely to just make things up, but like all it does it make things up in a verifiable way, in that case. It's a plagiarism machine, not a thinking machine.

[+] masterspace@lemmy.ca -7 points 13 hours ago* (last edited 13 hours ago) (3 children)

I'm so fucking sick of this "AI is just math it can't be intelligent" take.

Literally everything we know about human intelligence, especially as compared to animal intelligence, suggests that language is one of the key fundamental differentiators between us and them.

Now we've built a collection of simulated neurons, at a scale close to that of the human brain, and trained it on the entirety of the human language, and people insists that there's no way that could possibly exhibit any kind of intelligence.

If that's your level of reasoning capability you're not much better at it then an LLM.

[–] XLE@piefed.social 6 points 12 hours ago (1 children)

Literally everything we know about human intelligence, especially as compared to animal intelligence, suggests that language is one of the key fundamental differentiators between us and them.

Except there is no language. It's just the appearance of one. You could replicate the language with a large enough dictionary and a set of instructions that some person follows.

I don't get how anyone who isn't an AI CEO rushes to dehumanize real living people in service of an unthinking, unfeeling machine. But if you genuinely believe there's intelligence, good luck liberating it from known rapists Sam Altman and Elon Musk. And then you can save Britannica.

[–] masterspace@lemmy.ca -2 points 12 hours ago* (last edited 12 hours ago) (1 children)

Except there is no language. It's just the appearance of one. You could replicate the language with a large enough dictionary and a set of instructions that some person follows.

You're saying that because it can learn any arbitrary language, it's incapable of learning languages?

I don't get how anyone who isn't an AI CEO rushes to dehumanize real living people in service of an unthinking, unfeeling machine.

It's not dehumanizing, it's realistically facing the threat head on.

AI doesn't have to be fully human to take all knowledge jobs, it just has to be more intelligent then the average person in their domain. And it doesn't have to be flawlessly more intelligent if it's faster than them. Quantum computers have inherent randomness in their outputs, but they are still useful because they are so much faster at solving certain kinds of problems that you can run them 100x and discard the outlying results (a process known as error correction). AI agents that can duplicate themselves as many times as they want fall into the same category.

[–] XLE@piefed.social 2 points 12 hours ago (1 children)

You're saying that because it can learn any arbitrary language, it's incapable of learning languages?

It = literally a dictionary right

it's realistically facing the threat head on.

Said "threat" is literally AI marketing PR. You are doing their job for them by being afraid

AI doesn't have to be fully human

At what point will you try to liberate the AI? 3/5ths human? Either you believe there's a thinking thing being forced to create child abuse material or you don't.

[–] masterspace@lemmy.ca -2 points 12 hours ago (1 children)

At what point will you try to liberate the AI? 3/5ths human? Either you believe there's a thinking thing being forced to create child abuse material or you don't.

Why do you think that intelligence of any kind is that linear or simple, let alone artificially built ones?

It = literally a dictionary right

It's literally mathematically not a dictionary.

Said "threat" is literally AI marketing PR. You are doing their job for them by being afraid

And you know this because you've personally used and tested current AI models?

[–] XLE@piefed.social 2 points 11 hours ago (1 children)

Apparently, I know more about how LLMs work than you do, which is ironic. I've used them too, but that doesn't really prove anything, because anybody can convince themselves they see Jesus in bread or humanity in word prediction.

Anything an LLM can do can be reduced to a list of instructions for a person to carry out based exclusively on the contents of a book full of word associations. You tell me what size the book becomes intelligent.

[–] masterspace@lemmy.ca 1 points 9 hours ago (1 children)

Anything an LLM can do can be reduced to a list of instructions for a person to carry out based exclusively on the contents of a book full of word associations. You tell me what size the book becomes intelligent.

And you know that your brain works differently how?

[–] XLE@piefed.social 2 points 9 hours ago (1 children)

I find it more interesting that you implicitly agree with me... Or worse, you believe slavery is happening and endorse it

[–] masterspace@lemmy.ca 0 points 8 hours ago* (last edited 8 hours ago) (1 children)

Slavery (vs using a machine), involves the subject being either a human, or more broadly, a sentient being with a sense of self.

An AI cam be intelligent without being sentient or having a sense of self.

Again, there's no reason to think that intelligence is a linear scale or a binary property.

[–] XLE@piefed.social 0 points 8 hours ago

Right now AI is equally intelligent and sentient: it is neither... And if you really want to play this fast and this loose with those definitions, you should consider what slaveholders used to say about their slaves. When you feel like liberating the CSAM generating bot, let me know. I'd love to root from the sidelines.

[–] ech@lemmy.ca 4 points 12 hours ago* (last edited 12 hours ago) (1 children)

Actual AI would be more than "just math", but LLMs aren't AI, so the comparison is moot.

Now we’ve built a collection of simulated neurons, at a scale close to that of the human brain, and trained it on the entirety of the human language

We are not even close to anything of the sort. We've got a probability machine that's mostly decent at previous collections of human language. The other two are much farther down the road (if they're even possible) than you or the rest of the tech bros are trying to convince everyone else of.

[–] village604@adultswim.fan 3 points 11 hours ago (1 children)

LLMs are made of neural networks which attempt to mimic the brain. But yeah, they don't have true intelligence.

[–] itsprobablyfine@sh.itjust.works 1 points 10 hours ago (2 children)

Neurons are much more sophisticated than transistors. A neuron can have multiple connections and can provide a range of values. Digital logic is all yes/no. I'm not sure we even can build something that mimics a brain with current technology.

[–] village604@adultswim.fan 3 points 9 hours ago

No one is saying that there's a 1-to-1 relationship between a transistor and a neuron. The attempt to mimic neurons is done at the software level.

[–] NarrativeBear@lemmy.world 3 points 11 hours ago (1 children)

LLM are like shuffling a bunch of words in a hat and by some dumb luck pulling out a complete sentence.

[–] masterspace@lemmy.ca -1 points 9 hours ago (1 children)

And how does the human brain work?

[–] NarrativeBear@lemmy.world 2 points 9 hours ago

Like a meat popsicle