i love you a little. haha april fools. i love you lots
Show & Tell
Lint Roller? I Barely Know Her
Keni
will byers stan first human second
taylor price
art blog(derogatory)
trying on a metaphor

pixel skylines
Cosmic Funnies
No title available

❣ Chile in a Photography ❣
Not today Justin
i don't do bad sauce passes
h
I'd rather be in outer space 🛸
DEAR READER
noise dept.
dirt enthusiast

祝日 / Permanent Vacation

Kiana Khansmith

seen from United States
seen from United States
seen from Australia
seen from Mexico
seen from Mexico
seen from Colombia

seen from Mexico

seen from United States
seen from United States

seen from United States
seen from United States

seen from Colombia
seen from United States

seen from United States

seen from Malaysia
seen from United States
seen from United States
seen from United States

seen from United States

seen from United States
@lackofsurprise
i love you a little. haha april fools. i love you lots
[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?
Tumblr March 2026 Blackouts
So we're all pissed at the new update as we should be and I've been seeing many people proposing blackouts, which is amazing! But all the dates are different and people might get confused at what's happening when, so I just want to organize every blackout (at least that I saw) in one place.
So far I saw six people with dates.
The earliest one, organized by @yourlocalfandomfriendo begins on March 18th and will last 48 hours.
This overlaps with a second proposed blackout by @veejiez for March 19th.
There is also one on the 20th proposed by @daysleftofsecondterm and another one on the same day from 6AM UTC to 6AM UTC on the 21st by @everythingwsnormalhere.
These three days are all very soon so not everyone may see them in time to participate, but if you are able to participate for any or all of these days, I highly encourage you do. Otherwise there are two more blackouts coming up:
The next one after these will be on March 24th as organized by @aroacesafeplaceforall who suggested doing 12 hours.
And the last one, which I personally have a lot of hope for as it's a major day for activity on Tumblr and a blackout then could be especially impactful: April 1st, as proposed by @darkwood-sleddog
There is also a discord set up by @yourlocalfandomfriendo and @aroacesafeplaceforall for anyone interested in joining in!
SO OVERALL, it may sound like a lot, but no one expects everyone to participate to every date here. But PLEASE try to participate in at least one or two of these, even if you feel it may not do much.
Typical strikes, the ones we hear about all the time, win by withholding their labour for consistent periods of time; that's the power people have at work because that's what's exploited.
For blackout strikes, we need to withhold our attention; the resource we own which is exploited through the selling of both advertisements and data.
My comparison of blackout strikes with regular strikes will be for a whole other post, but for the time being, just know that
withholding our attention is our digital bargaining weapon
Tumblr literally lost 63% of its monthly traffic from 2024 to 2025; they are not in a position to play around with those of us still here.
So PLEASE try participating. We cannot let every decent online space get enshittified with no care or consideration for the communities using those spaces.
And where labour strikers risk losing incomes and jobs, all blackout strikers risk is... gaining some of their attention back for a little bit.
It Begins
And the ultimate bloging begins
It Ends
And the ultimate bloging ends
It Resumes
And the ultimate bloging resumes
I’ve had tumblr for 4 years but some of you bitches have had it for a decade. It’s time to seek penance
wait I’m curious now . Reblog this with how long u’ve been on tumblr for. Dating back to ur oldest blog ever !!!
The full performance of “I Lied to You” from “SINNERS” at the Oscars 2026
why does this have 32k notes? it’s just a picture of a knife in a ranch bottle, is there some unspoken joke that 32 thousand people share? what is going on here, i dont get it. it’s just a fucking picture of a knife in a ranch bottle. is there some spiritual connection people have to this picture? is there some ominous and mystical reasoning that this has 32 thousand notes? do people reblog this because it makes them look like some indie blogger? or is there just something funny to this? someone please explain
no one tell him
This is it, lads. The post that started us on this path 9 years ago.
I sure hope no one told him.
straight up stabbin “it”. and by “it”, i mean julius caesar
et tu brute?
Knives! Get your Knives here for no particular reason!
🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪🔪
Get em while they're cold, get em while they're sharp!
Special discount if your name is Brutus for no reason in particular!
Happy (late) international women's day to all the beautiful girls on tumblr
Barry McGlashan (British, 1974) - Lakehouse (2025)
Lygia Pape, Tecelar (Weaving), 1957.
(photo- Projeto Lygia Pape)
Virginia Woolf, Orlando: A Biography (first published on 11 October 1928)
ao3 come back
AO3 plz
Oxy aka Whatever Oxy (No Info) - Snowy Nights Just Feel Different, Photography