Shadow Wasser

I'm a nature lover nerd fanfic writer and other stuff. Hi.

[cinemamind]

Killer Klowns From Outer Space✨💫

[justalurkr]

Text of tweet under the cut because it is loooong.

But... Stochastic Parrots.

Timnit Gebru was fired from Google in December 2020 for refusing to retract a research paper, and every single warning that paper made about large language models has now happened at a scale the industry spent 4 years trying to make people forget about.

Her name is Timnit Gebru.

She co-led the Ethical AI team at Google. She co-wrote a paper called "On the Dangers of Stochastic Parrots" with Emily Bender at the University of Washington and two other researchers. The paper was 14 pages long. It was submitted to a top AI ethics conference. And it was the reason Google decided that one of the most senior Black women in AI research could no longer work there.

The story Google told publicly was that she resigned. The story she told, confirmed by 2,695 of her colleagues in an open letter, was that she was fired by email while on vacation because she refused to either retract the paper or remove her name from it.

The paper had not even been published yet.

Here is what she actually wrote, and why every prediction inside it has now come true.

The first warning was about scale itself. Bender and Gebru argued that training ever-larger models on ever-larger scrapes of the internet would produce systems that appeared fluent but had no actual understanding of language. They called these systems stochastic parrots because they would repeat patterns from training data with statistical confidence and zero comprehension. The paper predicted that this apparent intelligence would fool both users and developers into trusting outputs that were structurally incapable of being reliable.

This was 2020. GPT-3 had just come out. The paper predicted the hallucination problem before anyone had a word for it.

The second warning was about bias amplification. The paper documented in detail that internet-scale training data contains systematic overrepresentation of dominant viewpoints and underrepresentation of marginalized ones. The models would not just absorb this bias. They would amplify it, because the optimization process rewards confident outputs, and confidence in language patterns tracks frequency in the training set.

The prediction was that hiring tools built on these models would discriminate against women. That healthcare triage tools would underperform on Black patients. That loan approval systems would entrench inequality while presenting their decisions as neutral algorithmic judgment.

Every one of those things has now been documented in deployment.

Amazon's hiring algorithm penalized resumes that contained the word "women" in any context. Healthcare risk scoring algorithms used by major US hospitals were found to systematically underestimate the medical needs of Black patients. Apple Card's credit algorithm gave wives credit lines 10x lower than their husbands for the same financial profile.

The third warning was about environmental cost. The paper calculated that training a single large language model produced emissions equivalent to the lifetime output of 5 cars. The prediction was that the race to scale would create an environmental footprint that would eventually rival entire industries.

In 2024, Google's emissions were up 48% from 2019, and the company explicitly blamed AI infrastructure. Microsoft's were up 29%, same reason. Both companies have now quietly abandoned the climate commitments they were publicly celebrating the year Gebru was fired.

The fourth warning was about documentation. The paper argued that the training datasets being assembled were too large for anyone to actually audit. Nobody at Google, OpenAI, Meta, or any other lab could tell you with confidence what was in the data their models were trained on. This was not a temporary problem to be solved later. It was a permanent feature of the approach.

In 2023, researchers discovered that the LAION-5B dataset, used to train Stable Diffusion and other major image models, contained thousands of images of child sexual abuse material. The companies that had trained on the dataset had no way of knowing. The paper predicted that category of failure 3 years before it was found.

The fifth warning was the one Google cared about most.

Bender and Gebru argued that the deployment of these systems would centralize linguistic and cultural power in the hands of the small number of companies that could afford to train them. The internet would become a place where the dominant voice was a statistical average of dominant voices, presented as a neutral assistant. Languages underrepresented in the training data would degrade over time as more web content was generated by these systems and fed back into the next training run.

This is now happening in real time. A 2024 study found that 57% of new web content in English is AI-generated or AI-assisted. Researchers studying low-resource languages have documented active degradation in translation quality, because the synthetic content fed back into training is itself worse in those languages.

The paper Google fired her for predicted the model collapse problem before model collapse had a name.

The mechanism behind why this all happened is the part of her work that nobody quotes.

Gebru's argument was not that AI is dangerous in some abstract sci-fi sense. Her argument was that AI is dangerous in a very specific structural sense. The technology was being built by a small group of researchers who shared similar backgrounds, worked at similar companies, and were rewarded for shipping products faster than competitors. The incentive structure made it impossible for safety, ethics, and bias concerns to slow anything down. Anyone inside the system who raised those concerns was either ignored, sidelined, or removed.

She was making that argument from inside Google.

Then Google proved her right by removing her.

The team Google had built to make sure their AI was safe was dismantled in 90 days because they did the job they had been hired to do. Margaret Mitchell, the other co-lead of the Ethical AI team, was fired two months after Gebru for searching through her own emails for evidence of how Gebru had been treated.

Gebru did not stop. She founded DAIR, the Distributed AI Research Institute, in 2021. The mission is to do AI research outside the control of the companies that have a financial interest in not hearing the answers.

Every prediction in the Stochastic Parrots paper has now been validated by deployment. Hallucinations are an industry-wide problem the largest labs cannot solve. Bias amplification has been documented in hiring, healthcare, lending, and criminal justice. Environmental costs are larger than entire small countries. Training data audits remain impossible. Model collapse is an active research crisis at every major lab.

The question worth sitting with is the one almost no one in the industry will say out loud.

Every researcher with the technical credibility to call out these problems watched what happened to her in December 2020 and made a calculation about their own career. The number of people willing to speak publicly about safety and ethics issues inside the major AI labs collapsed after that firing and has not recovered.

The researcher Google fired for warning about exactly what is now happening was right.

The company that fired her is now the second-largest deployer of the technology she warned about.

And the people inside that company who agree with her are not allowed to say so.

[mbari-blog]

Serving some major squeeee 😱🥹🥰⁠

This adorable baby is the eryoneicus larval stage of a deep-sea lobster. These fuzzy globs of cute float and feed in the water column until they're ready to change into juvenile lobsters and settle on the seafloor. During this stage, they are covered with thousands of tiny hairs, which they eventually lose as they develop into adults. This larva is about five centimeters (two inches) long. You're welcome. 😍

[peregrinethegryphon]

I made something cursed but hopefully helpful to artists.

[crabdominalpain]

Brachiosaurus Monster that evolved in the deep dark caves, all its bones are cartilage and every part of it is prehensile to fit in crevices.

it climbs along walls, and is quite as a mouse.

[jaubaius]

The curious dance moves of the Striped Cuckoo.

[chick-it-out]

[heyyoufriendthere]

This new fish species is called Snuffleupagus. You can probably guess why | CBC Radio

S. snuffleupagus, a newly described species of fish, is named after the beloved Sesame Street character, Mr. Snuffleupagus, to which it bear

SNUFFLEUPAGUS REAL

[brightnessrandom]

Fantastic article!! The guys looking for it were fish researchers who saw it one time, knew instantly it was an undescribed species, and then tried for nearly 20 years to find and document it!

It's a type of ghost pipefish, related to seahorses, and it floats around coral reefs looking like a piece of algae and hunting unsuspecting prey

They are, of course, named after Snuffleupagus from Sesame Street!

Later on it the project, they got citizen science involved, and people across the Pacific started reporting sightings of snuffy fish from all over!

Hooray for science and hooray for S. snuffleupagus !

[paleoart]

Would you consider drawing pliosaurs

I've done a few!

[paleoart]

I just realized that apparently I never shared this series here on Tumblr??  

I call it Living Paintings, as they are recreations of extinct animals that some artists actually saw and painted from memory, a long time ago.

Patreon • Ko-fi • Facebook  • Twitter • Prints & Merch      

[paleoart]

Recent updates to this series: Glyptodon and Thylacine!

[paleoart]

OK SO a frozen baby Homotherium mummy has been found in Siberian permafrost and everybody is going nuts (including me). I had to make another installmente for this series with a figurine that has long been debated to either depict either a lion or Homotherium; Despite anatomical details pointing to the latter, there was no record of surviving Homotherium from the time the figurine was dated to have been made, so it was mostly believed to have been a cave lion. However, the frozen cub closes that temporal gap and makes it more likely that this was in fact the subject of the sculpture. So even though it’s not a painting, here it is brought to life.

[paleoart]

Two new updates:

A remake of the first ever piece I did for the series, the Megaloceros, given that although the piece of cave art I chose is often labeled as being that species, a more critical look to the antlers in particular makes me think it’s a red deer instead. I picked a more agreed-upon depiction of Megaloceros and also took the opportunity to make it a little more anatomically accurate.

Also, birb

[paleoart]

Some funky looking bugs 'round here 🐜

Available as prints and other merch formats!

Patreon • Ko-fi • BlueSky  • Instagram • Prints & Merch  

[mbari-blog]

This deep-sea fish has one of the best strategies for hiding in the deep-sea 🖤✨️

The California slickhead (Alepocephalus sp.) is often found below 1,000 meters (about 3,300 feet). Their name was inspired by the lack of scales on the fish's head. Those large eyes give the slickhead an edge in an environment where food can be scarce. It can glimpse even the faintest flickers of bioluminescence—the “living light” produced by deep-sea animals.

Deep-sea animals have a variety of remarkable adaptations to help them hide in the midnight zone, but this species uses the shade of their skin to hide from predators and prey in this dark expanse. Ultra-black fishes have unique structures in their skin that very efficiently trap and absorb light. Melanin—the same pigment found in human skin—is densely packed into super-thin layers on the outermost surface of their skin. While most light photons are immediately absorbed, the specific shape, size, and configuration of these melanin layers scatter any missed photons into neighboring skin cells, where they are subsequently absorbed. Ultimately, ultra-black skin absorbs 99.5 percent (or more) of the visible light with virtually none reflected.

[lecanerd]

In the ruins of the former American Empire, while travelling on dark routes, you may find yourself in the presence of an entity known only as "The bunny of the swarming bugs" or more commonly, "The Wabbit." He is said to be a long forgotten trickster god. Never quarrel with Him.

[twiststreet]

She got the idea for the study while walking with her advisor at Stanford to discuss her thesis topic, and the paper she eventually published in the Journal of Experimental Psychology in 2014 is sharp enough that it should have ended the seated meeting on the day it came out.

She ran 4 experiments on 176 people. Same person tested twice. Once sitting, once walking. The creativity tasks were the standard ones psychologists have used for decades to measure how good a brain is at generating novel useful ideas.

81% of participants in the first experiment produced more creative ideas while walking than while sitting. In the second experiment, 88%. In the third, 100%. Every single person walked into a more creative version of themselves. On average, people generated 60% more novel useful ideas the moment their legs started moving.

The skeptical question is the obvious one. Maybe it was the fresh air. Maybe it was the scenery passing by. Maybe it was the change of environment doing the work, not the walking itself.

Oppezzo killed every one of those explanations with one experimental decision. She put people on a treadmill facing a blank wall. No scenery. No fresh air. No environmental change. Just legs moving in place while staring at white drywall. The 60% boost held.

Then she ran the experiment that closed the case completely. She took participants outside in two conditions. Half of them walked through a Stanford courtyard. The other half were pushed through the exact same courtyard in a wheelchair. Same outdoor stimulation. Same scenery passing at the same speed. The only difference was whether the legs were moving.

The walkers produced dramatically more novel high-quality ideas than the wheelchair group. The outdoors did almost nothing on its own. The walking did everything.

She also tested the opposite kind of thinking. Convergent thinking. The kind where there is one right answer and you have to narrow down to it. Word puzzles where 3 words share a hidden fourth word that connects them. The seated participants did slightly better on these. Walkers got slightly worse.

Walking is not a general intelligence enhancer. It does one specific thing. It opens up the divergent search inside your brain. The part that generates options. The part that produces unexpected connections. The part that takes a problem and finds five ways into it instead of one.

When you need to converge on the single right answer, sit down. When you need to find the answer in the first place, get up.

The mechanism is now well understood. Walking selectively activates what neuroscientists call the default mode network, the system inside your brain that runs when you are not consciously focused on anything. The DMN is where mind-wandering happens. Where memories cross-reference each other. Where ideas that have been sitting in separate folders inside your head finally bump into each other.

When you sit at a desk and force yourself to concentrate, you suppress the DMN. When you walk at a natural pace, the executive part of your brain gets just busy enough handling the walking that the DMN comes online and starts doing the work that focus was blocking.

The most useful finding in the entire paper is the one almost nobody quotes. The boost did not turn off the moment people stopped walking. Participants who walked first and then sat back down stayed elevated. Their next round of seated creativity work was still significantly better than people who had been sitting the whole time. The rest lingered for at least several minutes after the legs stopped moving.

You do not need to do creative work while walking. You need to walk before the creative work. The brain holds the state.

Edited down a long tweet. (x)

[ivylaughed]

Link to the PDF of the study, found on the APA website.

https://www.apa.org/pubs/journals/releases/xlm-a0036577.pdf

[eclectickittysheep]

Ball Lightning Very Rare & Very Dangerous To People!

[misterlemonztenth]

misterlemonztenth.tumblr.com/archive

[hypostulatedoutcomes]

There isn't any definitive assessment of what ball lightning actually is, or how dangerous it is. This Wikipedia article might be helpful.

Ball lightning - Wikipedia

[3liza]

oh she is proceeding

[shadow-wasser]

Oh damn... big if real. I don't think *anyone* has *ever* filmed ball lightning before... lmk if I'm wrong. If this is real, it should be front page on like, Nat Geo.

[todaysbird]

discovery of the night

[gnomeknight1990]

Also quite possibly the cutest snail ever

[iamnmbr3]

Tom Riddle: I don't like having a first name that's so common. There are a lot of Toms.

Sorting Hat: No problem! I’ll put you in Slytherin. No Toms there... Because they’re all old blood purist families named after stars and things. Also they’ll hate you. Have fun sticking out and being immediately identifiable as everything they despise due to your unusual Muggle name. :)

[iamnmbr3]

@aloe-wasser Not unpopular with me! I've been meaning to write a meta about this. The personal qualities the Houses prize can be used for good or evil and do not (or at the very least should not) be taken in and of themselves to imply a certain moral value.

I think Harry actually epitomizes the good side of the ideal Slytherin traits. He is extremely cunning and uses trickery to avoid fighting when possible and to accomplish good deeds and his desire to take the initiative to save others in impossible situations and to defeat Voldemort himself exemplify selfless ambition.

Tom Riddle meanwhile is actually the perfect Gryffindor. His willingness to do his own fighting and act on his own and to start a revolution and his desire to face dangerous opponents in single combat rather than trying to get his minions to do it as well as the way he pushes magic beyond its known limits in multiple ways, including going through the painful and risky process of splitting his own soul many times when it's never been done before - all certainly qualify as daring. Harry will fight when he has to, but it's not his first recourse.

Riddle however is much more likely to treat charging into battle as a first option. His insistence on dueling face to face according to proper rules - both in book 4 and in book 5 when he reveals himself to Harry first rather than cursing him while his back is turned - even count towards the Gryffindor value of chivalry.

I think it's also notable that just as Harry had a reason to ask not to be in Slytherin, and thus ended up in the House that was the next best match, Riddle had a reason to ask not to be in Gryffindor. After his disastrous first interaction with Dumbledore I'm sure the very last thing Tom wanted was to have him as his head of House. Given his bloodline and how Gryffindor and Slytherin prized some similar traits, Slytherin is the next best choice.

Which would've been pretty miserable for him given that he was poor, had a Muggle name, came from a Muggle orphanage, and would have been presumed Muggleborn. Though even if he had the last name Gaunt the likes of the Blacks and the Malfoys and the Lestranges wouldn't be that impressed given that the Gaunts lived in utter squalor and had no wealth or power or status to speak of. And of course they would have considered Merope a blood traitor. So really he would hardly have been better off if his true lineage were known.

So things would not have been great for him. At least until he probably (in true Gryffindor fashion) won everyone's respect or at least tolerance by confronting them and starting a fight to force them to change their behavior.

Wild that Dumbledore just let him get sorted into the 'wealthy elitist Purebloods who hate Muggleborns and disdain those without wealth' House and was like "yeah it'll be fine."

[iamnmbr3]

Tom is pathologically unable to run away from a fight - even at the end of book 7, when he goes into having been very clearly told by Harry that it will get him killed. Harry isn't a coward but he has no problem running from a fight if it seems like the smart thing to do. He'd much rather use cunning to trick or sneak or talk his way out of trouble and although he certainly has a temper that can get the better of him, in threatening situations he tends to prefer fighting as a last resort and prefers avoidance. Tom on the other hand is more of the "can I fight it? no? I'm going to fight it anyway" mindset although he can certainly also be crafty.

[soup-of-the-daisies]

Harry and Tom mirroring each other is one of my Roman Empires. They both would’ve done well in their respective houses. It’s absolutely why I don’t subscribe to the (once common) idea that Harry’s Slytherin traits come from the Horcrux—if anything, Tom’s impulsivity and distaste for cowardice would’ve made Gryffindor the bigger contender.

It’d be fun to see a fic that makes Tom think, “Not Gryffindor, not Gryffindor”, as he’s being sorted… after which the Hat asks, “Are you certain? Gryffindor could lead you to glory.” Makes me wonder how his story would’ve changed, had he been sorted in a different house.

[iamnmbr3]

#to be absolutely clear i don’t think getting sorted slytherin results in automatic dark lordiness#but i do think it had a hand in whatever extremism (genuine or acted) tom riddle fell into#what route would he have chosen to become immortal? would he have made a friend who dragged him to their home during the bombings?#would his fear of death have been less paralysing?#who know via @soup-of-the-daisies

yeah yeah all of this. i can't help but feel that being in the 'monied purebloods who hate muggles, muggleborns and poor people' house can't have been a great way for tom riddle - an impoverished, apparently muggleborn orphan with a muggle name - to start out. (not that i think the other Houses were that much better when it came to these attitudes; just look at James Potter and how he treats Snape, or even his friends - he definitely sees Sirius, the wealthy pureblood, as more of an equal, even if it's not entirely conscious).

but yeah i lowkey headcanon that tom realized dumbledore was head of gryffindor house and became absolutely determined not to end up there. the reason he got sorted so fast was bc the second the hat went on his head he thought 'not gryffindor. not gryffindor or i will burn you' as hard as he could.

[shadow-wasser]

Hey y'all if you want an AU where Riddle is sorted into Gryffindor, give this fanfic a read; that's the entire premise:

There is Nothing to Fear

[shadow-wasser]

The Cull Block Tango

And now, the six adorabloodthirsty trolls of the Dream Bubbles,

In their rendition of: ‘The Cull Block Tango!’

Weak! Flip. Buzz. Ribbit. Video. Peixes.

Weak! Flip. Buzz. Ribbit. Video. Peixes.

Weak! Flip. Buzz. Ribbit. Video. Peixes.

Weak! Flip. Buzz. Ribbit. Video. Peixes.

They had it comin, they had it comin!

They only had themselves to blame!

If you’d have been there, if you’d have seen it!

I betcha you would have done the same!

Weak! Flip. Buzz. Ribbit. Video. Peixes!

You know how people have these weird personalities that get you down? Like… Taaaaaaaavros! Tavros was so pathetic. Not just pathetic. Weak! Well, I was chatting with him a8out how I’ve 8een playing the humans, and I’m REALLY good at it, And there’s Tavros, pretending he’s a hero, With his shiny new metal legs…….. no, not strong ones. Weak! So I said to him, I said, “Tavros, if you come after me I’ll give you a free hit!” And he did. So I let him charge and I turned his lance away from me........ Into! His! Chest!

He had it comin, he had it comin!

He only had himself to blame!

1 M3T VR1SK4 S3RK3T TH3 G4MB3L1GN4NT 4BOUT THR33 SW33PS 4GO 4ND SH3 TOLD M3 SH3 L1K3D JUST1C3 4ND W3 H1T 1T OFF R1GHT 4W4Y. SO, W3 ST4RT3D FL4RP1NG TOG3TH3R. SH3’D B3 TH3 CLOUD3R, W3’D PL4Y, 1’D K1LL 4LL TH3 3N3M13S, W3 SH4R3D TR34SUR3. TH3N 1 FOUND OUT. JUST1C3 SH3 TOLD M3! JUST1C3 MY 4SS! NOT ONLY W4S SH3 K1LL1NG 1NNOC3NTS, OH NO, SH3 W4S F33D1NG TH3M TO H3R LUSUS. ON3 OF THOS3 SP1D3RS, YOU KNOW? SO TH4T ON3 T1M3, ON TH3 M3T3OR, 1 4SK3D H3R 4BOUT JUST1C3, 4S USU4L. YOU KNOW SOM3 TROLLS JUST C4N’T GU3SS 4 CO1N FL1P!

She had it comin', she had it comin' (Weak! Flip. Buzz. Ribbit. Video. Peixes.)

She took a wiggler in her prime!

(Weak! Flip. Buzz. Ribbit. Video. Peixes.)

And then she used you (Weak!); and she abused you (Flip!)

It was a murder, but not a crime!

(Buzz. Ribbit. Video. Peixes!)

Now I’m Walking Around The Meteor, Hunting For Clowns And Fish, And In Storms Mr Ampora In An Absolute State. “i already killed you!” He Said. And He Kept On Saying, “but i already killed you!” Then He Ran Into My Chainsaw. He Ran Into My Chainsaw And He Fell In Half.

If you’d have been there, if you’d have seen it!

I betcha you would have done the same!

彼女が自分の罪を償う時が来たのだ。死と死、嘘と嘘。友人が障害を負ったことが最後の一撃でした。私のしたことは完全に正当化されました。

Yes, But Did You Do It?

ribbit ribbit

mY gOoD bRo EqUiUs HaD tHiS aDoRaBlE mOiRaIl. aNd ThE tHrEe Of Us HaD bEeN hAtEfRiEnDs FoR a BuNcHa SwEePs. fOr ThE lAsT pArT oF tHe GaMe I’d BeEn DoInG aLrIgHt In ThE fIgHtS aNd AlL. iMpS, oGrEs, LiChEs, GiClOpS, bAsIlIsKs, BlAcK kInG, oNe RiGhT aFtEr ThE oThEr. wElL aFtEr AlL tHaT wAs OvEr We WeRe AlL oN tHiS lOnElY mEtEoR, tHe TwElVe Of Us, TrOlLiNg AnD mAkInG sOmE nEw FrIeNdS. aNd I rUn OuT oF sOpOr, WhIlE i’M tRoLlInG tHiS aLiEn Tg. hE sEnDs Me A vIdEo, I oPeN iT. aNd ThErE’s My MiRtHfUl MeSsIaHs, A pArOdY oF tHeM bOtH, HeReSy! wElL, i WaS iN sUcH a StAtE oF sHoCk, I cOmPlEtElY bLaCkEd OuT, i CaN’t ReMeMbEr A tHiNg. It wasn’t until later, when i was cutting their heads off to preserve ‘em, THAT I EVEN KNEW THEY WERE DEAD!

They had it coming!

They had it coming!

They had it coming all along!

I didn’t do it,

But if I’d done it!

How could you tell me that I was wrong?

i lovved feferi peixes more than i could possibly say. she wwas a real empathic gal, sensitivve, an aquarist. but she wwas alwways trying to do good in the wworld, she’d go out all the time doing good in the wworld, and on the wway she did sollux motherfucking captor. i guess you could say wwe broke up because of quadrant incompatibility. i wwanted her flush. and she preferred to die.

That dirty bum, bum, bum, bum, bum!

That dirty bum, bum, bum, bum, bum!

I had to cull’ em! I had to cull ‘em!

They had it comin' all along!

'Cause if they used us, and they attacked us,

How could they tell us that we were wrong?

They had it comin' , they had it comin,'

They only had themselves to blame!

If you'd have been there, if you'd have seen it,

I betcha you would have done the same

“Tavros, if you come after me-”

“JUST1C3 MY 4SS! “

“Fell In Half…”

嘘と嘘

“A pArOdY oF tHeM bOtH,”

“quadrant incompatibility.”

I betcha you would have done the same!