spicy boi blogatorium yeehaw

he/himbo I wanted to do a Tumblr a while ago but then people would think I came from Twitter so now I'm doing a Tumblr as a Reddit refugee. some blorbos to expect: games, my cat, the state of Minnesota, bass, yelling at twitch, the simpsons, old man stuff

[bleeding-r0se]

beautiful harmony on my dashboard

[adamtotscomix]

[animentality]

[esoteric-merit]

This is about as unrelated as it gets, but I was once on a cooking competition show as background. I had a prep list, and while the contestants were front-stage, I was in the back of the shot, doing the prep list, looking chef-y, alongside some other BG. The prep list was double-duty, it was also making stuff for the contestants to use later, but our main job was just to make the kitchen look busier. I imagine that in most shots I was in, I was at best an out-of-focus blur behind the main shot. So at the beginning, the contestants are showing off their knife skills and making little fruit sculptures or whatever, IDGAF, and they were being interviewed at the same time. "What does cooking mean to you?" and that kind of stuff. I . . . was not supposed to be interviewed. I was BG. The interviewer apparently, did not know who was a contestant and who was BG, so they came over to me. And after all of these eloquent long-winded philosophy-and-dream guided answers, they come to me, and go "What is cooking?" And me, deep in focus on looking professional and getting the prep list done and not listening to the interviews, aware that I'm not supposed to talk or be talked to (I'm not paid for speaking lines!), but also aware I am the focus right now, but also aware that they will cut this out of the televised part and I can relax and shrug him off, but also suddenly aware there's still a live audience and I can't just shrug him off, but also confused because I didn't realize he was fishing for a deep philosophical answer: "Cooking is making food". Well. I got a laugh out of the live audience, and they must have decided to keep the line because my pay was upgraded from "SSE" (Special Skills Extra) rates to "Speaking" rates.

[templatelord]

best answer

[thegaymertrainer]

TikTok live is such a waste of time there are no good….

[cognitohazardous]

this is how we should generate power for cities

[transmechanicus]

contextual storytelling from that sign in the back

[quendergeer]

muttering "the hamsters control the wheels" as i wander around like a blind medieval soothsayer

[ankle-beez]

[qiyra]

[allie-leth]

@lost-estradiographer you have absolutely filled my dash with Chappel Gnome. I hope you are happy with yourself.

[xpxssxm]

Thrilled.

[waspsinyouryard]

@athena-turns-into-an-eagle you have absolutely filled my dash with Chappel Gnome. I hope you are happy with yourself.

[morelike-agayna]

Ecstatic.

[la-principessa-nuova]

well what we really need is a gnomognognomegnome

[morelike-agayna]

a what?

[la-principessa-nuova]

A GNOMOGNOGNOMEGNOME!!!!!!!

[icaruswaffles]

Is it chappel gnome day already? I forgot to get a present for her 😰

[desolendate]

im literally in tears what the fuck

[cat-mermaid]

was speeding up this footage to make a gif, but the audio is fucking golden, so its staying a video

[xpxssxm]

[parawyrm]

More from that one post

[parawyrm]

[parawyrm]

[parawyrm]

[icaruswaffles]

The Facebook post the Winona PD made had its comments turned off after people mocked and slammed them for crying over the $3k price tag for each camera. How about you just don't replace them and stop spying on your citizens.

[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?

[rederiswrites]

An article confirming and elaborating on the above.

[mothric]

them: why do you watch anime

me:

[clean-acts-for-reasonable-price]

[shadow-wizard-moneygang]

the abilities of the average tenor saxophone player

[identifying-saxophones-in-posts]

Actually that's an alto

[icaruswaffles]

"Somehow he's able to synchronize-"

BLAH BLAH BLAH that's actually just what happens when you play some bitchin jazz

[armengoldira]

[chuckletons]

just made the best non-looping gif i think

[chuckletons]

[chuckletons]

i said it was non-looping i'm sorry what am i supposed to say!!!!!!!

[redtextedquotes]

Everybody looking at this post

[yeahokayillreblogthat]

yeah okay ill reblog that

[reallyreallyreallytrying]

you have two posts in a row about placebos. why

one of them actually isn't about a placebo. that's probably the one you're having a reaction to

[wosp-reactivated]

Humans evolved cuteness aggression as a trait because cute things deserve to die

[fishing-princess]

For thousands of years humanity has selectively bred chickens to have larger breasts and to have to lay eggs constantly. We did it to get more meat and per chicken, but from the perspective of the domesticated chickens they’re sorta living in a very specific fetish scenario

[icaruswaffles]

Domesticated chicken is the level of bimbofication that Kirsti Noem's husband wishes to achieve