An Assortment of Hobbies and Interests

Random tidbits from a teacher, crafter, and overall nerd. Expect to see a lot of mushroom pictures, cosplay, and assorted fan art. INFJ | Asexual | Any pronouns are fine.

[chipper-pessimist]

🍉 Charity Commissions 🍉

Given the dire humanitarian situation in Gaza, I am opening commissions to raise money for a variety of aid and other organizations.

Here's how it works:

1) Choose what type of drawing you would like:

$10 = one character, bust (no background)

$20 = one character, full body (no background)

$40 = two full-body characters (no background) or one full-body character with background

These prices cover line art only; full color (done with COPIC markers) costs an additional $5 per character

2) Donate the corresponding amount of money to one of the following organizations (rated 97% or above by Charity Navigator): 

The Palestine Children’s Relief Fund

Middle East Children’s Alliance

Islamic Relief USA

ActionAid USA

Heart to Heart International

Doctors Without Borders

3) DM me with a screenshot of your receipt to confirm your donation, and a description of what you would like me to draw (reference pictures are helpful but not required).  I will draw portraits, original characters, and fan art.  I will NOT draw NSFW, gore, or anything that spreads hate.  Note -- DO NOT PAY ME DIRECTLY.  Your donation to one of the charities listed above counts as payment.

4) When the commission is finished, it will be scanned and sent to you as a digital file.  I can also mail hard copies of commissions if a mailing address is provided.

Here are some samples of my work:

[manuciruja]

awww the like button turns into a rainbow when you press it! that's so cute...hey staff what's with all the trans women you keep nuking?

[manuciruja]

i think we should be ridiculing them more for this. you don't get to try and go all "queer website" when your staff likes to go on nuking sprees targeting the trans fem users

[smadsel]

would be remiss not to mention that the rainbow notably straight up just removed the trans flag colors from it. like they’re gone. it’s the progress flag minus the trans flag colors.

[smadsel]

that’s not the whole flag, now is it

[babyanimalgifs]

Little fish eats his foods 

(Source)

[bunigojade]

[aflo]

this is so sad he doesn’t even know there’s a double barreled shotgun pointed at him

[identifying-fish-in-posts]

Pacific spiny lumpsucker (Eumicrotremus orbis)

[duke-of-fire]

[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.

[fatcatissurprisinglysvelte]

The paper is available online: https://dl.acm.org/doi/abs/10.1145/3442188.3445922

[higgsbison]

happy glorious 25th of may

[lizardho]

I’ve been thinking a lot lately about some of the people I interact with. I have a coworker who I am pretty sure is a MAGA type, and she is also a lovely woman who is dreadfully overworked and so good at connecting to patients when they call. I can see the conflict on her face when she talks to me, a gigantic tranny dork who speaks Spanish and affirms the LGBT community, but can also talk to her about her cows and knows about guns and stuff. I can see the fear in the eyes of my former Young Men’s leader when he misgenders me and realizes that I’m not an ideology but a person he has known for a long time. I can see the way my extended family stop and stutter over political discussions when they realize they are talking about me. And I don’t know why but lately it’s just made me think about my neighbor as a kid.

When we moved to Arizona, we moved next door to a lovely retired couple - John and Lucy. John was a veteran of WWII, he had an M.D. and a Ph.D. in radiology, and he LOVED us to pieces. His wife, Lucy, was a sharp and gifted woman - well spoken, very observant, and VERY clever. I just know that she used that cleverness as a mom to great effect, because with my and my siblings she always managed to find a way to send us home with candy and treats for a week despite my dad’s protests. We loved them, growing up, and even though they have long-since passed away I love them still, and I love what I learned from them.

John was, as stated, a WWII veteran. He was enlisted as a rifleman, and later as a front line medic, starting at Point Du Hoc and moving inwards to France and towards the Rhine. He let me do a report on him in 6th grade where he shared war stories with me he had kept to himself his whole life - he said it was out of respect for his friends who didn’t get to come home and tell their stories.

He said he told me because he knew I could respect the memories of his friends.

He showed me his collection of medals, and which he’d kept hidden away in a sock in his attic because he’d feel an immense grief any time he saw them. He had wanted to be a doctor his whole life, prior to being drafted he was studying medicine and had taken the Hippocratic oath to Do No Harm. He saw his medals as a reminder that he had Done Harm.

After telling me his stories he was able to convince himself that while he had Done Harm, it was only because his only other alternative was, to him, cowardice. He chose to be brave even if it meant acting against his Oath because he felt that if he didn’t do it someone else would have to go in his place and he would be responsible for the harm that befell them. I don’t think that’s true, but for him it was and that was something no being on earth could have ever dissuaded him from believing.

He shared wild stories - melee combat on the beach, clearing artillery bunkers, receiving a Purple Heart for being injured in hand-to-hand combat with a Wehrmacht rifleman he said he felt pity for because they were the same age and he had to imagine the man he was fighting had been drafted just like him.

He shared how he was awarded a Silver Star for charging a machine gun nest, but shared that he was most proud of not killing anyone in the process. He threw a grenade with the pin still in it and when the machine gunners jumped to avoid being blown up they were killed by someone else so he didn’t have to do it. He took the machine gun and shot the other machine gun in that French field to pieces so he didn’t have to kill the people operating it. He said they were giving out Silver Stars like candy but I knew he was being modest.

He told me about being redesignated as a medic, about how he crawled for about 500 yards on his belly to rescue an injured tank driver, then threw him over his back and crawled the same 500 yards back (1000 yards total) to treat his injuries. He said he met the man in an Army hospital in England after his spine was broken by a high explosive panzer shell was fired through a hollowed out French farmhouse and landed about 20 feet away from him.

He told me about all the people he helped and saved as a medic, he told me about his work in radiology and research after the war. He showed me a hallway that was quite literally wallpapered with academic honors he’d earned as a researcher. He told me about how his first Fourth of July back was a horror show for him because fireworks and German artillery make very similar sounds. He told me about how he woke up in a cold sweat well over half a century later hearing the screams of German artillery men being burned alive with flamethrowers, or hearing his own voice apologizing to the young German soldier he stabbed in the heart at Point Du Hoc.

He told me that when he was asked to present at a medical conference in Germany 25 years after the war ended that he was so scared he couldn’t step off the plane, and that his wife had to hold his hand and lead/pull him with her. He said he was not scared because he was worried about being triggered, but because he knew that someone somewhere outside of that plane had the course of their life irreparably altered by his military service. That to someone out there he was the cause of immense suffering and harm. That some unwitting waiter could be the son of the Nazi Officer he stabbed in the heart with a 12-inch hunting knife. That some woman asking questions in the audience would be the daughter or widow of a man he sent to judgement with a .30-06. He was scared that they would hate him.

He knew what the Nazi’s had done, he knew better than anyone I’d ever met. He’d watched the documentaries, he’s seen the PoWs returning from camps, he’d seen the civilians massacred and tortured by their regime, but he also knew that among the monsters were people like him - idealistic 20-somethings who only wanted to make the world better and were ripped away from that life by the Nazi war machine. And he spent his whole life mourning the loss of innocence and peace that was forced on so many people by such a corrupt power.

To be honest I don’t know if I could do that, but he could. He told me he could still feel the dead and lost with him, both when he slept and when he woke. He told me he thought he’d go to his grave never having told a word of this to anyone. That the stories of him and his friends and allies would disappear silently with him and those like him. That he had wanted that until he realized that he didn’t have to sell out to share the stories - that he could give the stories away for free to someone who would love the people in them, and not just the content of them. He didn’t want his stories to be used as Patriotic Pornography by some TV network or magazine. He wanted the people he knew to be respected, he wanted their memories to be honored and loved, and he entrusted me, a 12-year-old “boy” to do that.

He told me for years afterwards that after telling me these stories that he slept better than he ever had. That by sharing the stories with someone who could hear Him over the din of victory and glory and honor and revisionistic history. Someone who could see the man in the story and not just see the plot of a battle being won. He wanted to be human, and he wanted the people he saw die to be human too - everyone, not just the people on his side. He wanted someone to see and to know the anguish of having to look someone in the eye as heartblood muddies the ground beneath them and hope that they understand that this was not an act of love or hatred but an act of desperation. To hope that you had just taken out One Of The Bad Ones instead of a medical student or a poet who had been drafted. He wanted me to see how hard he had worked since then to build a world without scarcity, to build a world of peace. He wanted me to know SO badly that the cost of violence, any violence, even necessary violence, is always ALWAYS paid by both parties involved.

I think about the rise of the new right wing - the new Nazi movement’s traction in politics, and I feel sad and scared - the world that Johnathan J Yobaggy, my neighbor, my friend, and my hero, worked SO hard to build is being done away with by people who do not understand the cost of the path they are entering. I can see brief moments of recognition in the eyes of some of the people I mentioned - The former young men’s president who immediately regrets misgendering me and hen he makes eye contact with me and sees Me staring back at him and not a faceless “ideology.” I can hear it in the voice of my uncle who quietly comes up to me to apologize for some homophobic comment he made absentmindedly. I can see it in the eyes of racists and sexists being interviewed on TV when they realize that they didn’t vote for a concept, they voted for a real thing. And honestly, I have mixed emotions about it. Because while I understand frustration with the status quo, the importance of basic human needs like affordable good and rent, and I know the fear that comes with feeling powerless, I also can’t help but grieve the endless wheel of history bringing us back to this God Damned Fucking Place again. I hope we can avoid this fate, not just for our sake but for the sake of everyone who has ever tried to make the world safer. For everyone who has ever tried to make up for human nature, for everyone who has ever placed themselves on the offering plate to protect others from the cruelty they know lies just under the surface of mankind’s tenuous grip on progress. I want SO badly for there to be a solution to this, for the people who idolize the Nazi party and the impact of fascism to see that the price of this path is paid in more than just blood but in soul. That they’re allowing themselves to be devoured too. I want for the centrists and the fence sitters and the idealists who want to “change it from the inside” to see how dangerous our politics have become. I want them to see that they’re losing the things that make them great in exchange for a security blanket that’s now become far far far too small to ever work for them again.

Safety found in the past is already gone, and safety found in the future is only as real as a daydream. That any ideology that promises that by “joining us now we’ll make things rough so we can make things safe in a decade” is a promise made by those who will not have to fight the battles they send you to.

I don’t know if America was ever really great, but as long as John was alive it felt great to me. There is no ideology that can replace a neighbor. No tax plan that can replace a friend. No grocery bill that can replace community and connection. No amount of budget cuts that can replace kindness. No amount of suffering from people I hate that will ever make more love. I don’t know how to make America great, but I know how to make my America great and it is not by selling out integrity and compassion and community and fucking humanity to make eggs and gas cheaper. It is by seeing and hearing the people around me. I’m not Mormon anymore, but I still know the value of mourning with those that mourn and comforting those that stand in need of comfort. I’m not Christian anymore but I still have Eyes That Can See and Ears That Can Hear. I want to make this all stop but I can’t stop the collective power of tens of millions of people so instead I listen to my MAGA coworker tell me about how sick her kid was last week. I make jokes with my Young Men’s leader. I hug my uncle. I let them see me fully, as a human and not an ideology. As a woman and not the concept of gender. As a whole person and not someone who can be easily summarized or boiled down into something short and quippy. And I let them know I can see them fully too, and I can see all their humanity as easily as they can see mine. I just have to hope that this works - that enough people can See and Hear the people in their lives who matter to them to bring them out of their personal world of forms and into the real world.

I am probably, honestly, just spiraling a little bit. I took my ADHD meds today and in addition to helping me focus they make me a little anxious so I doubt things are as bad right now as they seem. But just in case there’s any truth to the way things seem to be going, remember, and I mean this seriously: Be kinder to each other, be gayer, and read more Terry Pratchett.

And for the love of god day hello to your neighbor.

[lizardho]

Johnathan J Yobaggy’s oldest son is, I believe, also a radiologist. I may have written this about your grandpa. If that is the case, I just hope you know how loved he is. If not, I still wish good things for you and your family because radiology is fresh to death.

[lizardho]

I got this from the talented @eagle-writes in the mail a week ago and I’m so touched at the effort that must have gone into this 🥰

[perplexingly]

My little 3d diorama, based on a painting by Józef Męcina-Krzesz, "Chopin's Last Chords", which I think is one of the gentlest, most beautiful paintings. Done in Maya, ZBrush and Substance Painter and rendered in Blender

Józef Męcina-Krzesz's original painting:

[a-totally-reasonable-username]

@namelessennes

[avidreaderffn]

@sandstonesunspear

[polyamorousmisanthrope]

Jesus Tapdancing Christ... THIS is a good welt pocket and the people who designed Simplicity 2895 ought to be blasted well ASHAMED of themselves for the crap way THEY wanted a welt pocket made. *SNARLS*

[makereadgrow]

This is how I learned to do it and a good example of what you want to see in a short form tutorial: pinning, pressing, seam finishing, good fabric handling.

I would mention that you can make the pocket facing with a small panel of your matching fabric that is visible and the rest in a lighter fabric to reduce bulk. That's a lot of denim layers for comfort.

[headspace-hotel]

USA Native Plant Resource Masterlist

Because Google is totally useless and won't help you with ANYTHING

iNaturalist: Take photos of living things you see, post them, and the community will identify them for you. Data from iNaturalist is used in scientific research.

Wildflower.org Plant Database: Enter search criteria and find some plants. Very useful if you're looking for plants with specific qualities or know what you have in mind.

Native Plant Finder: This website is still in beta and is a work in progress, but it will show you plants for your area ranked by the number of butterflies that use them for their caterpillars.

WildflowerSearch: AMAZING resource for identification and for learning about new plants. Shows you where plants are native/not native, TONS of search filters.

Native Plant Trust: A New England organization, but probably useful to anyone.

Northern Forest Atlas: Great images and identification resources for trees; has good pictures of bark, seeds, buds, leaves.

FloraFinder: Another plant database site that's being slowly built up by a passionate nerd.

MonarchWatch milkweed by USA ecoregion: Tells you what milkweed species you should plant for monarch butterflies.

Native Beeology: Not plants, but a closely related subject.

I will add more and post an updated list as I find more.

[sandersstudies]

My Shakespeare students (they are 12) wanted to summarize the lessons they learned this semester. If. Um. Anybody would like to see.

[sandersstudies]

I cannot emphasize enough that they made these with very little input from me.

Henry the Fifth

- ALWAYS encourage others to do their best.

- NEVER talk about people behind their back.

Antony and Cleopatra

- ALWAYS check your produce for pests. [They liked this one so much made a rap about it.]

- NEVER count your chickens before they hatch.

Hamlet

- ALWAYS act decisively

- NEVER tell your girlfriend to go to a convent and become a nun [Oh boy they REALLY liked this one]

Romeo and Juliet

- ALWAYS collect all the important information before making an important decision

- NEVER bite your thumb at us, sir. [They enacted this scene in the original language a lot, except they swapped every “sir” for “bro.”]

The Merchant of Venice

- ALWAYS pay your debts.

- NEVER judge based on appearances, because “all that glisters is not gold.”

The Tempest

- ALWAYS try to forgive others.

- NEVER be a colonizer. [Yes, a middle schooler said this]

Midsummer Night’s Dream

- ALWAYS stay on forest trails

- NEVER fall in love with an ass. [They were excited about this one for obvious reasons.]

Twelfth Night

- ALWAYS stay in touch with those important to us

- NEVER read other people’s mail

Macbeth

- ALWAYS wash your hands. [One of the girls performed Lady Macbeth’s entire Out Damn Spot monologue at the end of the semester]

- NEVER succumb to peer pressure.

[sandersstudies]

Yeah, I was re-reading the Tempest like “hmmm will they even understand the subtle themes here… this might be a cut-and-dry magic story to them.”

Kid 1 (known intellectual): Wait, Prospero is like… a colonizer to the magical creatures. He showed up on their island and enslaved them.

Kid 2: Enslaving people is bad! Is Prospero a bad guy?

Kid 3: But Caliban is bad! He wanted to kidnap Miranda.

Me: Yeah, it’s kind of hard, isn’t it? Just like how in real life most people are a mix of good and bad.

Kid 4: …is this why Shakespeare is supposed to be, like, really good?

[moniquill]

[dannyvantass]

This Alaskan parka is windproof, waterproof — and made from bear intestines | CBC News

It's waterproof. It's windproof. It's lightweight and durable. And it's made from the intestines of two bears, painstakingly cleaned and sew

I really wanted to know more about this, especially how the water proof stitching works. Here's more information on this project, and hopefully more in the future!

[tehnakki]

This is amazing! I need to make a note to play around with that water proof stitch technology soon.

[nocturnal-stims]

Flexible feather armor

🪽 Miscreations_us on IG

[j-train13]

Astronauts are so funny man. Here's just a couple of things I've found hilarious from this past week of space stuff:

It's probably already been spread around here enough already, but in case anyone's missed it; 7 hours after launch, commander Reid Wiseman, dealing with tech issues, uttered the generational quote "I have two Microsoft Outlooks and neither one of those are working."

After fixing the issues that were afflicting the onboard toilet, mission specialist Christina Koch (who has quickly become my favourite of the four) laughingly said “I’m the space plumber, I’m proud to call myself the space plumber.”

On Easter Sunday, the Artemis II crew hosted a makeshift egg hunt, by hiding packets of dehydrated scrambled eggs around their Orion capsule.

The way the crew always makes sure to make it very clear they're in space when doing interviews. From stuff like Wiseman just hanging out floating sideways on screen or Koch letting her hair loose so it can freely span out flowing around her.

While in transit, the crew decided to record a parody of those bad 80s sitcom intros where everyone turns and smiles at the camera.

When the crew reached the furthest point from Earth in the mission, they jokingly clambored over each other in an effort to get to the far side of the capsule, so that they could individually claim to be the furthest person from earth.

At the same time, on the ISS which was at the time on the other side of earth, the 7 astronauts onboard had a light-hearted race to the far side of the station, making jokes about being the furthest humans from Artemis.

On the way back to earth, NASA actually managed to establish an audio call between the crews of the ISS and Artemis II (where they shared the above info), and Koch called one member of the ISS crew, Jessica Meir, her "astro-sister" as the two of them previously spacewalker together in 2019. Meir then responded I'm so happy that we are back in space together, even if we are a few miles apart" (a few here being 230,000).

While Jeremy Hansen was doing an interview, Wiseman and Koch were just in the background swatting the mission mascot (a little moon plush toy named Rise) back and forth between each other.

[shutyourmoustache]

[kimyoonmiauthor]

US Book ban on LGBTQIA books

The Nationwide Book Ban Bill Moves to the House: How to Take Action Now

The nationwide book ban bill, HR 7661, has progressed out of committee and into the House. Here's what you need to know to take action.

At least hit the shiny repost button if you're going to do nothing else. This should not stand.

Stop the bill dead in its tracks. LGBTQIA people have a right to be represented in books.

[daysleftofsecondterm]

1,034 days left

[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?

[undeadcorvid]

Let me save anyone who's curious a step: Here's the paper on ResearchGate

[the-beard-of-edward-teach]

Hawaiʻi is currently in the midst of a natural disaster if you didnt know

Apparently there isn’t much news coverage of this outside of the islands

Towns are flooded, homes destroyed and collapsed, roads collapsed, lives at risk, gas leaks from the flood damage

Haleiwa and Waialua are currently evacuated because the 120 year old dam is at risk of bursting

Mind you that damn is owned by Dole. Theyve known about it needing to be fixed for years and years and years. Despite having more than enough money they refuse

The state has been trying to buy it out from them for years so they can fix it, but the sale hasn’t gone through

Keep in mind that the Dole family were the ones who illegally imprisoned Queen Liliuʻokalani and illegally overthrew the monarchy.

If I see another goddamn person say how sad this is for the tourists whose “trips were ruined” and compare a messed up vacation to people losing their homes, belongings, and livelihoods, I’m going to lose my mind

I am so lucky that my family or friend’s are safe and the few whose houses flooded didnt have it too bad, but so so so many were not as fortunate

If you haven’t heard anything about this until now, I suggest looking into it

The sirens didn’t go off until the flood had been going on for hours. Our state government is spending so much money on a fucking monorail we don’t need rather than fixing the infrastructure.

It’s been the locals and Kanaka doing the most to help get people to safety from the start

I don’t really know how to end this

I just need to know people are aware

I need to know people are seeing whats happening