I'm like what if agender ace aro was a lesbian
noise dept.
h
No title available
Mike Driver
DEAR READER
wallacepolsom

roma★

shark vs the universe

★
Aqua Utopia|海の底で記憶を紡ぐ
taylor price

@theartofmadeline
tumblr dot com
Game of Thrones Daily
AnasAbdin
ojovivo
Misplaced Lens Cap

Origami Around
Keni
Sweet Seals For You, Always
seen from United States
seen from Canada
seen from United States
seen from Chile
seen from Austria

seen from Romania

seen from Germany
seen from Poland
seen from United Kingdom
seen from United States
seen from United States
seen from United States
seen from Netherlands

seen from Indonesia

seen from United States

seen from United States

seen from United States
seen from Malaysia

seen from Brazil
seen from United States
@esiako
I'm like what if agender ace aro was a lesbian
[source]
🚨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?
Link to the cited paper
Like students facing hard exam questions, large language models sometimes guess when uncertain, producing plausible yet incorrect statements
What are they looking at?
fuck, marry, kill: the wound that won’t heal, the past you can’t undo, the ghost that keeps returning
How to make leaf bouquet for spring
strikingly blessed and all that
i love susuie
My favorite human ritual is the unspoken rule that if you enjoyed a concert, you must clap without stopping at the end (if you were seated, you may rise to express further respect). The musician will bow, then exit the stage, but you must keep clapping. The musician must return and act surprised, bow again, then exit once more, and you continue to clap. Then, the musician will return and play one or two extra numbers (you stop clapping during the music) and at the end after they leave for the last time, you can clap as long as you wish but the musician will not return. It’s just such a cute song and dance. I've been to shows where the musician expects it (to the point that i could see their timer backstage that indicated how much time they had left for the show and they bowed for the first time with 20 minutes to spare) so they just go through the motion of pretending to end the show but the extra number is completely planned and we all expect it. Everyone in the audience is in on it but we all just do it anyway because it’s like a conversation were the audience and the musician are saying Thank You to each other over and over. Makes me feel some type of way
When the pairing is so good that it doesn’t matter if you see it as romantic or platonic. What matters is that you see them and know that one is not complete without the other. Frequently bought together. Do not separate. If you separate them one of them will bite you
ONE ART IMPROVEMENT!!! Here’s some things I noticed… head/face shape became rounder and more consistent (which i think is the most noticeable change), eyes slightly bigger, noses also seem bigger+more 3D, bodies more proportional, hair neater, lines smoother…
These are comparing faces from the earliest and latest chapters, but his art improved/changed a lot during the middle too of course…
I just… really had to add Reigen to this post because I feel like he’s the most noticeable example of improvement and I’m just so proud of ONE.
This is important I really wanna explain
There’s so much pressure to be better-than-great if you want to pursue any kind of career. And that kind of hyper-perfectionist attitude burns out so so many people who would otherwise want to try something but don’t feel worthy of even attempting.
And you know, ONE’s art right now isn’t typical manga quality, but it’s good. It’s good and it’s so amazing to me because of how visibly it’s improved. ONE is someone who started with drawing skills you see on the left and said “fuck it” and wrote manga because it’s something he wanted to do. And he’s improved so much through constant practice.
I’ve seen posts talking about the professional manga remake that One Punch Man has, as well as the brilliant animation that go into the OPM and MP100 animes, and those are all incredible things. But it seems to brush over the fact that ONE is still the one doing the front-running. He’s still putting out the manga chapters for OPM and MP100.
I love the success and recognition ONE is getting. That’s why I know about him to start. But all the brilliant and beautiful adaptations aside, I love seeing reminders that ONE himself is still the one making his own content, still improving, and still putting out an incredible story.
Microsoft Productivity Pack for Windows (1992)
Why do the computer and printer have a complicated relationship
have you met a printer
the ship of theseus wikipedia article in 2003. 20 years later, after 1792 total edits, 0% of its original phrasing remains. (x)
And yet it's still the same article, the article designated to be about theseus' ship. Debate solved i guess
easy as fuck . were they stupid?
so a very long time ago, my dad worked with an arson investigator
this guy was often one of the first people on the scene following a suspected arson, once emergency services had done what they needed to do. at times, there were also civilians on the periphery. often, they were freaking out, and understandably so; their home or workplace had just, quite literally, gone up in smoke
this investigator wouldn’t try to calm them down. he wouldn’t comfort them or be a shoulder to cry on.
instead, he’d walk up to the person most visibly losing their shit, hand them a fire extinguisher, and say “hey, can you keep an eye out for any other fires, and if you see one, can you put it out with this?”
of course, there was no actual risk of another fire. he wouldn’t be on the scene investigating if there was even a chance that the fire wasn’t completely put out. but the bystander didn’t need to know that
because that person, without fail, would immediately pull it together, take the fire extinguisher, and stand guard. they were, at least temporarily, calm enough for this investigator to do this job
my dad has told me the parable of the fire extinguisher a hundred times, and i think about it a lot. i think about what it says about people and crises. i think about what it says about the grounding power of having a purpose. and i think about the importance of letting someone help me through something, even if that help is just going to be another casserole to throw into the freezer, because useless or not, that fire extinguisher might be the only thing holding them together
Every bit of emergency response training I've ever done has a part that's like "check the bystanders, and if anyone looks like they're about to lose it, give them something to monitor. Tell them to keep an eye out for the ambulance or to make sure everyone else is staying clear of the hazard or something."
I once sat through a social psychology lecture about the bystander effect and (ignoring that it turns out there were hundreds of calls to the emergency line during the initial incident that made people coin the term, so people weren't doing "nothing",) you know the best way to break people out of being bystanders?
Give them a job.
My lecturer had a chance to test this- she was on the scene for a motorbike crash, realised everyone had frozen and so started giving orders "you, call an ambulance, you and you, stand there so cars know to go around etc etc."
And it worked. Of course it worked.
Most people are dying to have a script to help. It's why part of first aid training is telling everyone: You. You are in charge until someone with more skills/training gets there. Step 1 call for help, step 2 make sure you won't get hurt helping, step 3 if you can apply first aid until help can get there.
You've got this, and if you don't got this at least you're trying to help.
funfunfun gang
* Won't you play again, Kris?
antlers
@demilypyro
OKAY here's all of them together. the gradual progression of me getting braver with her design since december is kinda funny.
thanks to those who suggested the first two hairstyles. took a lot of liberties with the third one lmao.
🐟💚