Oh my fucking god, Thaisha fucking Lloy. Wreathed in storm, surrounded by her people, priestess of her city, telling this foreign false faith to fuck right off before they get hurt. Oh my god.
Like, the difference between the Thaisha of the ritual, the gentle shepherdess guiding lost souls through grief and joy to freedom, and then this Thaisha Lloy, stone cold, stone faced, just flatly daring anyone to touch what is hers. No explanations, no nothing. Just flat, this is the way it is because I say so. Do we have a problem?
I am getting Delenn of the Minbari flashbacks, because Thaisha Lloy just made a whole combat just go away. Tail between their legs.
There's all these Magpies in the background with their crossbows, Vaelus, Azune, Hal, Murray, half of damned Dol Makjar ready to have a fucking go, and none of it is needed.
Just Thaisha. Just a Lloy. Casually ready to stand her ground.
Brennan is very good at using NPCs to support PC development, choices, and story. Hannan (rip) is a great example.
First, Brennan uses Hannan as an antagonistic force toward the Seekers; he unifies them by standing against both Occtis and Vaelus. He also provides contrast to the more measured and peaceful Druids we've seen thus far. He provides a counterpoint to Vaelus, the only other Elf we've seen in the campaign, and forces her to reveal more about her position regarding the Shaper's War by defending it.
As his relationship with Vaelus develops, we get more nuanced glimpses of what Elvish life is actually like, the full weight of Sylandri's hold of her creations, and Vaelus admits who she actual has been missing and grieving all these years; not Sylandri, but her family. Admitting this lets us understand the sacrifice she makes when she chooses to put her family aside, to break the Stone of Nightsong and help the Rungjani dead find peace.
And of course, when Hannan learns of Mara's capture he flies to Obrimus Manor to help her. There, he does not barge in to search, instead he starts to counter the Undead. This is extremely important support for Thimble. The magical darkness dissipates, her search difficulty decreases significantly, and the undead threat is lessened greatly. I feel confident in saying that Thimble would not have succeeded without this intervention, but critically, he supports Thimble's success without swooping in to make it his own.
With this support, she is able to finish off the Undead Monstrosity that was Raimond Davinos. Now Teor's death is not in vain because she heroically finished the work he had started. Together, they rid the world of a great evil while freeing Raimond's spirit and recovering his blade.
Thimble is able to recover the Pridesires. Their spirits and bodies cannot be harnessed by the Tachonises. She recovers the Royce Bracers and the Einfassen ax, which could have complicated things had they been discovered in the crypt. She also finds valuable information throughout her search, learning where the Gnomish statues were taken and information on the tenebral ship. Hannan facilitates all of this as he holds the darkness and the dead at bay.
The last important thing Hannan does is to die.
And this is very important, narratively.
Primus Tachonis has been bested by the work of the Magpies. We've seen him threatened by the Einfasen and condescended to by Yanessa Halovar. Why should we be threatened by him? Maybe he's just prideful and sloppy. Maybe he's careless and stupid. Maybe he's actually not that frightening.
And then he kills Hannan without a second thought, and we see the truth that Brennan has been nudging us toward since the beginning.
Primus is not stupid or short-sighted, he is panicking.
He knows things we do not. He knows something is coming, and the full wrath of the Druids that may come in a handful of months doesn't matter, because there are larger forces at play. So yes, the Einfasen and the Cormorays are deeply powerful and the Halovar's have power and influence as well as cunning and poise the others seem to lack.
Even so, Primus Tachonis is the most dangerous threat right now, because he is desperate. In his mind, he has nothing to lose. The murder of Hannan proves this.
Sometimes the cost of a great story is the loss of a great character.
You know, I was so caught up in the shock of Hannan dying and Wick almost getting the sequel to that that I didn't have time to think about how amazing it is that Thimble got Teor, Cyd, the torn banner, the bracers, AND the greataxe out of that house.
the idea that Thjazi is hiding from House Tachonis in Julien's shadow is so fucking funny, not even just for the whole beef Julien has with him, but because Thjazi was known as Shadow
I think this has been said to death already but goddamn Aabria is so locked in on the thematic resonance of this campaign it’s crazy, any time Brennan touches on something Thaisha is connected to, Aabria picks it up immediately and spins it into something beautiful,,, her druid magic and calling, the Circle of Ancients, the Lloy family, the Pariah Blades, the treatment of the orcs and their liberation, Thaisha’s feelings about the Shapers and the War,,, she uses it all to cut so incisively to the emotional core of these big themes
It’s such good character creation because Thaisha isn’t just connected to the history of Araman, its personal to her it’s all so personal to her,, Aabria hasn’t just written an interesting factual history, she’s seen some of the questions and conflicts at the core of this worldstate and made a character who has deeply held beliefs and strong emotions in answer to those questions and it lets her make character choices that both make sense for Thaisha as an individual and connect to the themes of the campaign at the same time and it’s so good it’s just so good Aabria Iyengar you’re very good at this we should be telling you this every day
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.
definitely read under the cut for the full overview
essentially, everything Gebru warned about LLMs in her paper has come to pass, including "model collapse," the issue where LLM datasets grow worse as more of their training data - the internet - becomes AI-generated slop, degrading the LLMs more and more as time goes on
because LLMs aren't really artificial intelligence and cannot understand anything - they're just really good at sounding like the dominant language - their "hallucinations" (nonsense and lies) will continue to worsen. they're not even really algorithms anymore, just "Stochastic Parrots"
the human, cultural, and environmental cost is awful and building. when these LLMs become unusable and all the hundreds of billions of dollars invested evaporate, the economic collapse will be unimaginable
Do any of u have decent recipes that are like 5 ingredients (not including spices) and take 45 mins or less to prepare i gotta stop eating sandwiches for dinner
ignore the title of this google doc because it's a long story but it's a really solid recipe for southwest chicken alfredo
this is a vegetarian potato curry recipe that's about 75% spices; once you get the potatoes in there you can really do whatever you want with it
this is literally just pasta, broccoli, and cheese babey and you can live off that shit for DAYS it makes such a big portion
bro this spinach/pesto/3 cheese flatbread is so fucking tasty bro
also you can make the flatbread yourself it's super quick!!
oh hey I'm eating this white chickpea chili right now, much like the curry it's mostly spices and you can do p much do whatever you want with it
don't let the name fool you these potatoes are delicious any time. not just breakfast.
this is slightly more than five ingredient when you add them together but if you have time and really wanna fuckin treat yourself I recommend these chicken strips + this cornbread + either these potatoes or these buttered veggies on the side.
this recipe for gogumabap (sweet potato and rice) saved my life when i couldn't eat hardly anything for a long time. the recipe itself calls for a heavy bottomed pot but you can absolutely use a rice cooker and put the rice and diced sweet potato in together and just let the machine do its thing
If you have a crockpot, I do a really good pulled pork carnitas that is:
Pork - any cut boneless
Onions- chopped anyway you like
Orange juice - about one cup (you can use about 2 tablespoons of the condensed frozen stuff if you have it)
Root beer, pepsi, or other cola soda - about one cup
Preferred chili mix seasoning - 1-2 packets depending on how intense you like your flavoring and how much pork you put in
Garlic - as much as you wanna, specifically smashed not cut
Canned or frozen corn - to your preference
Dump all that into your crockpot in the morning (or the night before) and let it run on low until dinner. Cook some rice if you feel like it, or just serve over tortillas.
the thing i keep thinking about is that azune probably never said anything in his whole life about his sister who he loves very very dearly and would really like to find please.
because like. I think thjazi could've found her. between thjazi and thimble, the number and reach of their contacts, I think they for sure both could and would have found that girl for him. at least have gotten word to reach her that her brother was alive and looking for her.
I think he never said anything, and they never knew there was a sister to find, and they both remained alone.
The crazy thing is, ninety-nine times out of a hundred, if you asked me on any given day "Would like to see a picture of some genitals?" my answer would be "😰 No, that's... No, thank you. I'm okay, actually." I have nothing but the utmost respect for people who do engage with the penis side of the internet, but personally, I've spent the better part of two decades doing all I can NOT to have pictures of dick and balls or sexy bikini babe buttcheeks blasted onto my retinas constantly. And yet... to be denied the penis? To have a jumped up pile of javascript tell me, a grown adult with an air fryer and an outstanding council tax bill, that I cannot be trusted to withstand the sight of a bare nipple unless I let it scan my drivers' license? I will move heaven and earth to see that fucking nipple, friend. I will walk a thousand miles barefoot on hot coals before I give you big brother bitches my passport number. A thousand miles through the desert with five VPNs just to press my face up against the glass and see the last uncensored picture of two My Little Pony Characters sixty-nining each other, and I don't even want! to look at it! But I will! I must! for the sake of our fucking democracy!
This took a hot second to make, but I really enjoyed this combat and the whole Schemer's Arc in general! Very glad the hounds are safe :,D
ALSO! I don't like asking for this sort of thing, but I'm trying to get onto VGen to start doing commissions, so any like or comment is greatly appreciated! <3
theartofmadeline
discoholicmusic
amandla
areudirty
stormborns
vaporwave
