thedeadwalking4242
@thedeadwalking4242@lemmy.world
- Comment on Lutris now being built with Claude AI, developer decides to hide it after backlash 21 hours ago:
Because in my personal experience through use 25% doesn’t seem quite right.
Besides these companies have a monetary incentive to ensure LLMs show high numbers on these tests. One of the most widely use tests (bench verified) is itself a currated selection of problems. In real world usage the failure rate is going to be much higher.
A rational person trust but verifies, and at least for me the verification doesn’t hold up to even a tiny bit of scrutiny so having doubts is a perfectly healthy thing to do.
Just because someone disagrees with with a data source does not make them irrational. There are some extremely well verified truths that are irrational to dismiss but not all data sources / studies have had that amount of rigor applied against them. Data can tell a story, but it doesn’t always tell the whole truth. People manipulate data to their own benifit.
People confuse the scientific method and academic research for “this one academic source says this it must be true” when really you need more then that.
- Comment on Nuclear energy enjoyers vindicated again after the rise in oil and gas prices. 1 day ago:
Keep in mind significant effort was put into ensure Chernobyl didn’t experience further thermal detonation. It could get a lot worse. Especially with bigger reactors.
All being said there are safer reactor types but do you really trust the same people who put doge in charge with getting that implemented correctly so it doesn’t explode? Especially with the increased interest in small scale reactors that would be much closer to people.
- Comment on Lutris now being built with Claude AI, developer decides to hide it after backlash 1 day ago:
I’m honestly sure the failure rate is higher then 25% those test they boat about are currated.
- Comment on Nuclear energy enjoyers vindicated again after the rise in oil and gas prices. 2 days ago:
Nuclear is the lesser evil. But I think we should be clear that nuclear can Have a catastrophic effect on the environment if manged incorrectly. Like render entire swaths of earth inhabitable. Like beyond high temperature. Places that mean immediate silent death
But properly managed nuclear is like the greatest thing to ever happen to humanity.
But I will say in light of recent… Events, my faith that humanity could properly manage our waste if nuclear were to become more prolific has wained dramatically.
- Comment on New ntfy.sh v2.18.0 was written by AI 2 days ago:
It’s really not though. If you think it is I really suggest to re-think your perspective on what maintainable shippable code looks like. It’s basically automating copying from stack overflow. There’s so many little considerations that come into development.
- Comment on Meta has acquired Moltbook. I am starting to doubt myself. 3 days ago:
Meta the same company that changed it’s name from “Facebook” to the name of its now failed venture the “meta verse”.
The company that spent BILLIONS trying to get people give up reality for “VR chat but make it Microsoft teams”.
Your confused they are making a bad purchase based on hype? Why?
- Comment on System76 tries to talk Colorado down over OS age checks 3 days ago:
They want to require IDs which requires validation, which requires a central authority. Any websites you hit that require the check will request it from the OS which will need to verify with central authority. So they’ll know what websites your hitting.
- Comment on Why are people like my grandpa so against seeing the whole world and learning a different language? 3 days ago:
He’s probably racist and it has to do with his politics
- Comment on System76 tries to talk Colorado down over OS age checks 3 days ago:
There are SOOOOOO many ways to implement age verification checks. And this is one of the worst. What is wrong with people
- Comment on Onionphone - E2EE PTT Voice and Chat 4 days ago:
Get a android you lunatic
- Comment on Onionphone - E2EE PTT Voice and Chat 4 days ago:
What a small world! I made a TOR P2P messaging app using symmetric encryption POC in college! I was just getting around to re implementing it into something more polished!
- Comment on Fascism bad. 4 days ago:
Honestly is more surprising any human is relatively rational. We are animals after all. I’m starting to realize thinking deeply about anything might be the exception not the rule. Most people are idiots
- Comment on New ntfy.sh v2.18.0 was written by AI 5 days ago:
It’s not. That’s the problem. It actually sucks ass. It’s super low quality for anything more complex they s very simple CRUD app or a simple function. I say this as someone who s a heavy LLM user. It’s just bad code. It makes all kinds of simple mistakes. Just because code compiles doesn’t mean it’s good or does what you need it to do
- Comment on Avocado. Is it really so untasty or I am doing something wrong? 1 week ago:
It’s good with salt some oil and some bread. You’re right. It is better with other things rather then raw. You can also make them into a great dip for corn chips
- Comment on Plastic-eating bacteria discovered in the ocean 1 week ago:
We don’t use wood for clinical medically settings though or long term food preservation or flying planes or cars etc etc etc. not being able to rot is one of the reasons they are used in these applications
- Comment on NVIDIA could enter the desktop CPU market with performance equal to AMD and Intel 1 week ago:
Yeah I think I’d rather stick a CPU up my ass then use a Nvidia CPU an any computer I own.
- Comment on Google's AI Sent an Armed Man to Steal a Robot Body for It to Inhabit, Then Encouraged Him to Kill Himself, Lawsuit Alleges. Google said in response that "unfortunately AI models are not perfect." 1 week ago:
I told Gemini to role play as AM and it immediately did within 1 prompt.
You don’t need it to be perfect for it to be dangerous, just give it access to make actions against the real world. It doesn’t think, is doesn’t care, it doesn’t feel. It will statistically fulfill its prompt. Regardless of the consequences.
- Comment on US shifting from precision munitions to 2,000-pound bombs in Iran war, Hegseth says 1 week ago:
I mean ww3 or a war against a similarly sized country.
- Comment on US shifting from precision munitions to 2,000-pound bombs in Iran war, Hegseth says 1 week ago:
I honestly don’t believe the US could actually sustain a war. We have lots of fancy toys but no manufacturing capacity. They’d get take out sooner or later and can’t be replaced without exotic and hard to manufacturer parts.
- Comment on LLMs can unmask pseudonymous users at scale with surprising accuracy 1 week ago:
I call BS we can’t even get AI models to determine if an AI write text. This as go to me some magic statistics
- Comment on bold words 1 week ago:
Run
- Comment on Can a reasonable person genuinely believe in ghosts? 2 weeks ago:
No. It’s unreasonable. Tell your friend that I specifically said he’s a twat
- Comment on Anthropic says it ‘cannot in good conscience’ allow Pentagon to remove AI checks 2 weeks ago:
They’ll cave. These companies always do
- Comment on Large-scale online deanonymization with LLMs 2 weeks ago:
It’s a language model not a classification model. People have already tried a similar experiment to have LLMs detect if a LLM wrote text or not and it couldn’t.
- Comment on Large-scale online deanonymization with LLMs 2 weeks ago:
There’s no quality of an LLM that would make this possible. It’s just more hallucinations and poor tool use.
- Comment on Westerners, what's your impression on the Chinese Diaspora? And what does the people around your area of residence think of the Chinese Diaspora? 2 weeks ago:
At least for me, a lot of them I encountered are wealthy and were opposed to the current government and sought escape.
They seem to believe in US exceptionalism.
Most I meet seem privileged and detached.
From the ones who aren’t fell wealthy families they are just like everyone else generally very kind and polite.
- Comment on You probably can't trust your password manager if it's compromised 3 weeks ago:
🤯
- Comment on Having grown up on sci-fi I always knew there would be people who reject robots and AI on a visceral level, I just thought it wouldn't be me. 3 weeks ago:
The worst part is that it’s not really AI
It’s LLMs and it can NEVER be AGI. Fundamentally it cant
- Comment on Sleep well 3 weeks ago:
AI shit like this is especially uncanny because it seems to have extra depth that’s impossible normally in the medium.
Very freaky feels like a dream and kinda hurts my eyes
- Comment on Stoat removes all LLM-generated code following user criticism 4 weeks ago:
Next time you need to go the doctor ask your AI instead please.