Tomjinks and Highfoolery

Sophie, Scotland born, Bisexual, Feminist. I blog whatever catches my fancy really. Main interests: Video Games, Books, Movies, Music, TV Shows, Social issues. Whatever else that catches my interest you may very well find on this blog. Well that's all for now :) have fun!

[important-animal-images]

[7777777awawawa7777777]

sorry everyone we won’t be seeing any men today they’ve all been bricked into their enclosure

[m1dori-eyes]

The Masc of Amontillado

[erinbowbooks]

[wizardshark]

99% of queer discourse stops right before they define the true difference between bisexual and pansexual!

[saspitite]

FOR THE LAST FUCKING TIME

BISEXUALS GROW FROM THE GROUND

PANSEXUALS GROW FROM THE CEILING

[importantcatpics]

[longseasons]

[gwydionmisha]

Bonus: If I buy a book I get to keep it! The publisher can't turn up at my house at random and confiscate all the books I bought.

[ismenes]

you know those studies showing that cursing helps with pain tolerance or whatever. that’s how i feel about making my weird little noises to get through my basic daily activities. sometimes you just have to go hggblaaaah for a minute so you can find the strength within yourself to get up or wash the dishes or send an email. mmmnneh. urgh. the torments are unending but you can always make some little sounds about it.

[only-cat-memes]

[traggalicious]

Wait this is literally a bomb ass drawing of that one post,,,

This one

I love it it’s perfect

[maybeasunflower]

My cat can open her jaw wider than I can open mine.

[only-cat-memes]

[hometoursandotherstuff]

https://www.bbc.com/news/articles/cwydx34kzlvo

[runcibility]

"Vanderhorst had been under the influence of MDMA and three litres of vodka she had consumed on the night of the offence last September, her lawyer Michael Hill told the court."

three. liters.

[anais-ninja-bitch]

i support women's wrongs

[homoqueerjewhobbit]

[only-cat-memes]

[sacred-portal]

[cargopantsman]

[nojamnojelly]

[gif87a-com]

[zukosgay]

is germany okay

[soloontherocks]

oh scheiße

[unscharf-an-den-raendern]

This isn’t quite how I imagined the second coming of Christ.

[not-a-slayer-of-dragons]

What makes this funnier is that I’m pretty sure that’s at the station for cologne cathedral

[i-dont-crow]

It is and everytime I see that hole I think of this video

[babtest]

Remembered to check at the station today and report that

The structural stigmata are still there

[keepcalmandcarriefischer]

Why are the 4 holes?? Maybe 5?

[official-german-puns]

It’s his daily commute.

[thatswhywelovegermany]

#it is written that twice did jesus stumble and twice did the cross strike nearby architecture #so really this is method acting

[ninatastic]

Unfortunately they repaired it recently

[pluto-rain-her]

Es hängen jetzt sogar Schilder dort:

In Kölsch, Deutsch und Englisch 😂

[belladonnaprice-deactivated2025]

[hollow-keys]

ID: A tiktok of a person saying "What's a word you pronounced incorrectly one time and it still haunts you to this day?" Justin Timberlake stitched the video and replied "Umm, me."

[millenniallust4death]

I’m paying to force seven thousand strangers to see a photo of my late husband having fun with his dog. Tumblr Blaze is totally worth it. XD

[millenniallust4death]

Thank-you to all of my new Internet stranger friends for being so gracious about having my post shoved onto your dashboards. I loved reading all of your kind tags and comments! Both Martin and Bosco have been gone for several years now but for 24 hours, they felt very present in my life. I greatly appreciate this gift. ❤️

[psycorax]

Reblog to have your dashboard be visited by the spirit of joy that death can end but not erase.

[leavescrown]

Love that this is well beyond 7000 people now and still going

[millenniallust4death]

@leavescrown Exactly! It’s a beautiful gift. Martin and Bosco out there travelling around the Tumblr community, continually making new friends.

[millenniallust4death]

@sseanettles

#hello again martin and bosco!! sending you boys round for another go :)

Reading your tag made me laugh out loud. It’s like two old friends unexpectedly stopped by your porch for a quick visit. XD

[petermorwood]

I’ll always reblog Martin and Bosco when they splash across my dash, because of Reasons.

What’s loved, lives.

[catasters]

[idontgettechnology]

The time of his life.

[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?