Technically Nerdy Moon @technicallynerdymoon - Tumblr Blog

sometimes being a fan of something means not wanting them to make any more of it

uma thurman should not have been in the movie. quynh should have been the big bad. the movie should have been about the emotional weight of andy and quynh reconnecting and andy's mortality that should have been contrasted with nile coming to terms with her new immortality and what that means for her and how that affects her family with the subplot of booker earning forgiveness and learning how to live with his ghosts and not just survive and they should have Talked About It, especially joe and nicky, who were undeniably the ones most hurt with his betrayal and uma thurman should not have been in the fucking movie

#the old guard #we could have had such a strong through line of quynh forgiving andy and the gang and the gang forgiving Booker but nooo #also fuck whatever they did to nile

non-un-topo

My Shaylas..... (x)

#the old guard #i miss my squad

petruchio

im literally always saying this

#my life summarized

dogpuppy

catharticscream

Remember in 2010 when Taio Cruz said "I throw my hands up in the air sometimes"? I appreciated his restraint. You can't just throw your hands up in the air whenever. There's a time and a place, and that time was 2010, and the place was the club.

#it is was one of of the deepest sadnesses of my life that I wasn't alive to through it back in the club to taio cruz #that being said I went hard to it during 5th grade dances... so I did what I could

local-lesbian-tree-demon

they should invent a body that feels normal to be inside of

I want a sky daddy screen adaptation purely because I want to see Linda's final vision board irl.

#kate folk #sky daddy #her face next to plane with “just married” took me out

klfg

[source]

klfg

🚨BREAKING: OpenAI published a paper proving that ChatGPT will always make things up.

Not sometimes. Not until the next update. Always. They proved it with math.

Even with perfect training data and unlimited computing power, AI models will still confidently tell you things that are completely false. This isn't a bug they're working on. It's baked into how these systems work at a fundamental level.

And their own numbers are brutal. OpenAI's o1 reasoning model hallucinates 16% of the time. Their newer o3 model? 33%. Their newest o4-mini? 48%. Nearly half of what their most recent model tells you could be fabricated. The "smarter" models are actually getting worse at telling the truth.

Here's why it can't be fixed. Language models work by predicting the next word based on probability. When they hit something uncertain, they don't pause. They don't flag it. They guess. And they guess with complete confidence, because that's exactly what they were trained to do.

The researchers looked at the 10 biggest AI benchmarks used to measure how good these models are. 9 out of 10 give the same score for saying "I don't know" as for giving a completely wrong answer: zero points. The entire testing system literally punishes honesty and rewards guessing.

So the AI learned the optimal strategy: always guess. Never admit uncertainty. Sound confident even when you're making it up.

OpenAI's proposed fix? Have ChatGPT say "I don't know" when it's unsure. Their own math shows this would mean roughly 30% of your questions get no answer. Imagine asking ChatGPT something three times out of ten and getting "I'm not confident enough to respond." Users would leave overnight. So the fix exists, but it would kill the product.

This isn't just OpenAI's problem. DeepMind and Tsinghua University independently reached the same conclusion. Three of the world's top AI labs, working separately, all agree: this is permanent.

Every time ChatGPT gives you an answer, ask yourself: is this real, or is it just a confident guess?

parkersloths

#trust fx #primo nizzuto #my beautiful evil bby boy

loreleywrites

im always saying this

feefal

Oyster mermaid~

feefal

ah fuck, so sorry ma’am-

annabelle--cane

stop. analyse that text through the lens of its author's intentions and original historical context. okay now take the author out back and kill them dead and analyse that text as though it were published by your mutual yesterday and is in direct conversation the contemporary discourse that's most relevant to your life. okay now pick your favorite angle of interpretation and come up with the strongest possible argument against it. now imagine that the text is your best friend and that it means you well and that you naturally give it every benefit of the doubt because you're on its side and you want the best for it. now imagine that the text wants you dead and it'll eat you if you don't eat it first. now pretend that you found this text locked away in a cave with no evidence of when or where it came from and you have to divine its meaning solely through its internal coherence and nothing else. okay now address the elephant in the room aspect of the text you've been ignoring because you find it boring or confusing or uncomfortable and become the number one expert on it. now spend forty minutes assigning all the characters dnd classes with at least three sentences of reasoning each. okay now do the cha cha slide.

heathercollinsmd

Josh Johnson in Charlotte (posted 4/21/26)

#this what I've been saying #the beauty has found its way to target men #its one of those thing that natural for a women to suffer through but God forbid a man go through it #anyway kill beauty standards uggos forever

myturtlespeedy

This is a result of the inhumane decisions that members of this administration want you to be silent about in public for fear of a loss of “civility”.