#corrupted nightshade's art

[corrupted-nightshade]

William doodles

just a bit of practice

{more below to shorten post}

___~___Bonus doodle___~___

he was drawn small, so she's crispy quality. 😔

___~___Info___~___

App used: Medibang

drawing tablet: XP-Pen

Tools used: G pen 2

Five Nights at Freddy's: Scott Cawthon (+ Steel wool)

Reference image for bottom right drawing:

#my art #corrupted nightshade's art #art #digital art #digital drawing #five nights at freddy's #fnaf #william afton #fnaf william #the purple guy #fnaf purple guy #burntrap #scraptrap #fnaf afton #fnaf security breach #security breach #fnaf 6 #fnaf pizza simulator #tw: body horror

[not-terezi-pyrope]

Often when I post an AI-neutral or AI-positive take on an anti-AI post I get blocked, so I wanted to make my own post to share my thoughts on "Nightshade", the new adversarial data poisoning attack that the Glaze people have come out with.

I've read the paper and here are my takeaways:

Firstly, this is not necessarily or primarily a tool for artists to "coat" their images like Glaze; in fact, Nightshade works best when applied to sort of carefully selected "archetypal" images, ideally ones that were already generated using generative AI using a prompt for the generic concept to be attacked (which is what the authors did in their paper). Also, the image has to be explicitly paired with a specific text caption optimized to have the most impact, which would make it pretty annoying for individual artists to deploy.

While the intent of Nightshade is to have maximum impact with minimal data poisoning, in order to attack a large model there would have to be many thousands of samples in the training data. Obviously if you have a webpage that you created specifically to host a massive gallery poisoned images, that can be fairly easily blacklisted, so you'd have to have a lot of patience and resources in order to hide these enough so they proliferate into the training datasets of major models.

The main use case for this as suggested by the authors is to protect specific copyrights. The example they use is that of Disney specifically releasing a lot of poisoned images of Mickey Mouse to prevent people generating art of him. As a large company like Disney would be more likely to have the resources to seed Nightshade images at scale, this sounds like the most plausible large scale use case for me, even if web artists could crowdsource some sort of similar generic campaign.

Either way, the optimal use case of "large organization repeatedly using generative AI models to create images, then running through another resource heavy AI model to corrupt them, then hiding them on the open web, to protect specific concepts and copyrights" doesn't sound like the big win for freedom of expression that people are going to pretend it is. This is the case for a lot of discussion around AI and I wish people would stop flagwaving for corporate copyright protections, but whatever.

The panic about AI resource use in terms of power/water is mostly bunk (AI training is done once per large model, and in terms of industrial production processes, using a single airliner flight's worth of carbon output for an industrial model that can then be used indefinitely to do useful work seems like a small fry in comparison to all the other nonsense that humanity wastes power on). However, given that deploying this at scale would be a huge compute sink, it's ironic to see anti-AI activists for that is a talking point hyping this up so much.

In terms of actual attack effectiveness; like Glaze, this once again relies on analysis of the feature space of current public models such as Stable Diffusion. This means that effectiveness is reduced on other models with differing architectures and training sets. However, also like Glaze, it looks like the overall "world feature space" that generative models fit to is generalisable enough that this attack will work across models.

That means that if this does get deployed at scale, it could definitely fuck with a lot of current systems. That said, once again, it'd likely have a bigger effect on indie and open source generation projects than the massive corporate monoliths who are probably working to secure proprietary data sets, like I believe Adobe Firefly did. I don't like how these attacks concentrate the power up.

The generalisation of the attack doesn't mean that this can't be defended against, but it does mean that you'd likely need to invest in bespoke measures; e.g. specifically training a detector on a large dataset of Nightshade poison in order to filter them out, spending more time and labour curating your input dataset, or designing radically different architectures that don't produce a comparably similar virtual feature space. I.e. the effect of this being used at scale wouldn't eliminate "AI art", but it could potentially cause a headache for people all around and limit accessibility for hobbyists (although presumably curated datasets would trickle down eventually).

All in all a bit of a dick move that will make things harder for people in general, but I suppose that's the point, and what people who want to deploy this at scale are aiming for. I suppose with public data scraping that sort of thing is fair game I guess.

Additionally, since making my first reply I've had a look at their website:

Used responsibly, Nightshade can help deter model trainers who disregard copyrights, opt-out lists, and do-not-scrape/robots.txt directives. It does not rely on the kindness of model trainers, but instead associates a small incremental price on each piece of data scraped and trained without authorization. Nightshade's goal is not to break models, but to increase the cost of training on unlicensed data, such that licensing images from their creators becomes a viable alternative.

Once again we see that the intended impact of Nightshade is not to eliminate generative AI but to make it infeasible for models to be created and trained by without a corporate money-bag to pay licensing fees for guaranteed clean data. I generally feel that this focuses power upwards and is overall a bad move. If anything, this sort of model, where only large corporations can create and control AI tools, will do nothing to help counter the economic displacement without worker protection that is the real issue with AI systems deployment, but will exacerbate the problem of the benefits of those systems being more constrained to said large corporations.

Kinda sucks how that gets pushed through by lying to small artists about the importance of copyright law for their own small-scale works (ignoring the fact that processing derived metadata from web images is pretty damn clearly a fair use application).

#not-terezi-speaks #ai #nightshade

[probablyasocialecologist]

A new tool lets artists add invisible changes to the pixels in their art before they upload it online so that if it’s scraped into an AI training set, it can cause the resulting model to break in chaotic and unpredictable ways.  The tool, called Nightshade, is intended as a way to fight back against AI companies that use artists’ work to train their models without the creator’s permission. Using it to “poison” this training data could damage future iterations of image-generating AI models, such as DALL-E, Midjourney, and Stable Diffusion, by rendering some of their outputs useless—dogs become cats, cars become cows, and so forth. MIT Technology Review got an exclusive preview of the research, which has been submitted for peer review at computer security conference Usenix.   

[...]

Zhao’s team also developed Glaze, a tool that allows artists to “mask” their own personal style to prevent it from being scraped by AI companies. It works in a similar way to Nightshade: by changing the pixels of images in subtle ways that are invisible to the human eye but manipulate machine-learning models to interpret the image as something different from what it actually shows.  The team intends to integrate Nightshade into Glaze, and artists can choose whether they want to use the data-poisoning tool or not. The team is also making Nightshade open source, which would allow others to tinker with it and make their own versions. The more people use it and make their own versions of it, the more powerful the tool becomes, Zhao says. The data sets for large AI models can consist of billions of images, so the more poisoned images can be scraped into the model, the more damage the technique will cause. 

[...]

Poisoned data samples can manipulate models into learning, for example, that images of hats are cakes, and images of handbags are toasters. The poisoned data is very difficult to remove, as it requires tech companies to painstakingly find and delete each corrupted sample. 

#nightshade #AI #ai artwork #AI art #artificial intelligence

[prisiidon]

okay so I tried using Nightshade and Glaze both together, and its hit or miss depending on the art. One artwork might be degraded pretty badly even on the lowest settings, while on another artwork the degradation is hardly noticeable. There needs to be more research on what affects degradation e.g textures.

Nightshade doesnt offer much protection against mimicry but helps corrupt GenAls as an offense. Glaze prevents style mimicry but it might be able to be bypassed so deglazers might be a thing 💀

#saw someones glazed art mimicked but with danny devitos face purposely added in rip #^i hope they're lying about the art they used being glazed sfdjksfd #fixating on this topic for the last 4 hrs :)))))))))))) #pri rambles #pri posts

[socksandbuttons]

From now on, run your art through Glaze and Nightshade. Glaze protects it from AI theft, and Nightshade poisons and corrupts the AI data pools, punishing any generator that steals it.

Im tired scoob

#would that even work for my art??????? #thats still YEARS of art already jsut out thereeeee #cant we just rebel or something #im gonna end up emailing cause um... Bruh #altho i think staff mentioned this too and yeah

[thequeenofthewinter]

Work-in-Progress Wednesday

Just because I am not writing long fic doesn’t mean I am not working on projects. Below you'll find my first ever WIP Wednesday art project, and after that have a little snippet of a ghost story prompt I am working on. (It's fresh out of the brain cells, so it may be slightly rough.)

Tagging: @mareenavee @oblivions-dawn @tallmatcha @changelingsandothernonsense @throughtrialbyfire @rainpebble3 @snowberry-crostata @paraparadigm @ladytanithia @wildhexe @dirty-bosmer @your-talos-is-problematic @umbracirrus @gilgamish @orfeoarte @skyrim-forever

Blue Dahlia

Untitled as of yet...

In the misty marshes of lowland Hjaamarch many stories are told—ones of strange magic and others of inexplicable mischief. However, as of late, many more are told of a much more insidious and unsettling nature. Chilling anecdotes have reemerged—ones which grandparents had passed from their grandparents before them in whispers of the wind long thought forgotten.

It is here. They are coming. 

At first they had all thought them to be old wives’ tales, stories which were told to small children to scare them into not coming home so late from playing in the marshes. Harmless. Safe. A good bit of fun to scare annoying younger brothers and sisters. But then, it started happening: the disappearances, the strange noises, and finally, the reappearances. Children which had been lost out on the marshes suddenly came back—or at least a part of them.

Everyone had ignored it as best they could at first, passing off the peculiar behavior as a phase or some innocent game that they were playing. However, over the days and weeks to come, the noises quickly became too loud to ignore and soon, Hjaalmarch was plunged into the darkest of nightmares—a twisted, corrupted quagmire of Vaermina’s greatest masterpiece. Only it was not her. 

No one and nothing—not Aedra, Daedra, nor anything from this plane to the next could save them from what was to come, as scarlet mushrooms cropped up out of nowhere and chilling vapors left behind the scent of deathbell and nightshade in every room. Only one common factor connected it all together: the children who had been whisked away for three days and three nights.

When any of the children were asked about where they were or what had happened to them, it was the only time they were oddly silent, almost catatonic as they looked with blank eyes and vacant stares upon their parents, caretakers, and eventually the Jarl herself.

“Jarl Idgrod, what are we to do?”

“Could you have not foreseen what was to come?”

“What is happening to our village?”

Questions but no answers appeared as more and more citizens brought their children to the longhouse. And here she had thought vampires were the peak of her problems.

#Skyrim fanfiction #dahlia #winter does art? #wip wednesday #winter writes

[0li0llie]

! IMPORTANT ! - ARTISTS & MEMESTERS PLEASE READ:

As we all know, with the advent of Generative AI, artists everywhere have been targeted in AI database scraping without consent- and with this, many creative industries are being abused by such companies. If you are an artist, art hobbyist, memester, or any type of content-creator, I urge you to try using Glaze and Nightshade on your works.

What is Glaze? "Glaze is a system designed to protect human artists by disrupting style mimicry." - Sand Lab, University of Chicago What does this mean?: It cloaks your image by adjusting inputs to pixels that are invisible to the human eye. So while the image may remain (minimally) unchanged to the human eye, generative AI will see something very different upon scraping it. It's almost like UV light, invisible to humans, but "visible" for AI. If you have more FAQs about Glaze, please read more here: https://glaze.cs.uchicago.edu/faq.html

"Oh but OliOllie, I only use tablet/phone/my pc is too old and tired for this program! I can't possibly use Glaze, it's inaccessible to the everyday internet user!"

Fortunately for you, the University of Chicago has this all figured out! Presenting WebGlaze: a wonderfully accessible way to cloak your images and join the fight against AI. Access is invite-only, but it is very easy to be invited! All you have to do is DM them on @TheGlazeProject on Xwitter or Instagram, or email them. Once you get an invite and create an account, just go to https://webglaze.cs.uchicago.edu for Glazing.

Images attached for comparison: unGlazed VS Glazed Kirby fanart by yours truly.

What is NightShade? "Nightshade works similarly as Glaze, but instead of a defense against style mimicry, it is designed as an offense tool to distort feature representations inside generative AI image models." - Sand Lab, University of Chicago What does this mean?: Much like Glaze, it cloaks your image by adjusting inputs to pixels that are invisible to the human eye. However, unlike Glaze, it doesn't just cloak your image to be hard for AI to read- it changes what AI sees entirely! An image of the Mona Lisa to you can appear as a picture of a little house on a hill for the AI, and enough poisoned images can tamper datasets to generate incorrect imagery. If you have more FAQs about NightShade, please read more here: https://nightshade.cs.uchicago.edu/whatis.html Nightshade is not yet available via WebGlaze at the time of me writing this post, but the Chicago team will be eventually making it an add-on to WebGlaze once they further study how the two methods coexist.

I remember seeing it mentioned somewhere that as little as 1000 poisoned images can corrupt AI datasets!

Ya'll better start using this! The more poisoned images there are out there, the better!

#art #artists on tumblr #digital artist #Artist #artistsoninstagram #traditional art #drawing #fanart #artwork #glaze #nightshade #no ai art #anti ai #digital art #concept art #enviroment art #character design #ALL CREATORS ON THIS DAMN PLATFORM #kirby

[ask-sibverse]

art asks! for any oc you'd like

pencil, marker, crayon, charcoal, tortillon, and creative! please

Oh my god holy shit okay! I'll do Mari and Belladonna as they're my two favorite OCs I have for this blog

Pencil (Does this character have any "sketchy" habits? Have they ever broken the law?)

Belladonna- I mean, his magic literally creates poisonous plants as a side effect, although that probably doesn't count. Possibly his preference for amplifying the worst negative traits in a person?

Mari- while I don't think they've actually broken the law, I feel like they'd be the type to graffiti, if given the chance

Marker (what's one thing your character would never tattoo on themselves, even if paid a million dollars?) (Jeez this is hard because I absolutely love getting tattoos)

Belladonna- probably anything that would remind him of his origins. Daisies (his Dream) and apple blossoms specifically. the opposite response, his ideal tattoo would probably be vines.

Mari- a cage or anything representing physical constraints. Her ideal tattoo would probably be a crescent moon or something symbolic of Nightmare

Crayon (what were they like when they were four years old?)

Belladonna- he was a very curious child who tended to stick close to his twin. Some of his curiosity is still there as an adult, but buried.

Mari- the events in her AU "muted" her personality quite a bit. She was a very energetic and talkative young child.

Charcoal (Do they have any unique physical traits that are not scars or tattoos?)

Belladonna- he literally has his namesake (atropa belladonna, deadly nightshade) growing from and winding around his bones.

Mari- the butterflies in her ref are attached to her body. They seem to be at least somewhat alive, too, as they flutter and occasionally move to different parts of her head if necessary to avoid something

Tortillion (How do they "blend in" with those around them?)

Belladonna- his "social skills" are often him mimicking his twin as his twin is "more liked". And that is not always a good thing.

Mari- when her SOUL is functional, she usually "blends in" by keeping a calm attitude and a pleasant smile (although she can privately be quite the mischievous gremlin)

Creative (free space for the author)

Aaaaaaaa holy shit this was a lot of fun! And honestly working on these two OCs are some of my favorite things. Belladonna is honestly so fun to "pick his brain" to speak. As a Nightmare, he has absolutely done horrible things. He almost wrecked his universe. And yet he has a "redemption arc" in sibverse at least that involves him becoming a better person sure, but does not regret what he did. (Also the corruption actually causes him physical agony constantly but that's its own issue)

And Mari is also a lot of fun for me as well. Sibverse has very few human characters, and Mari has been through a lot. And goes through even more in the fic. (Or fics... Theres enough content I'll want to split it up into three, possibly four fics)

#sibverse #dreamtale #nightmare sans #dreamtale nightmare #undertale au #frisk undertale

[pers-books]

ARTIFICIAL INTELLIGENCE

This new data poisoning tool lets artists fight back against generative AI

The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. 

By Melissa Heikkiläarchive October 23, 2023

October 23, 2023

A new tool lets artists add invisible changes to the pixels in their art before they upload it online so that if it’s scraped into an AI training set, it can cause the resulting model to break in chaotic and unpredictable ways. 

The tool, called Nightshade, is intended as a way to fight back against AI companies that use artists’ work to train their models without the creator’s permission. Using it to “poison” this training data could damage future iterations of image-generating AI models, such as DALL-E, Midjourney, and Stable Diffusion, by rendering some of their outputs useless—dogs become cats, cars become cows, and so forth. MIT Technology Review got an exclusive preview of the research, which has been submitted for peer review at computer security conference Usenix.   

AI companies such as OpenAI, Meta, Google, and Stability AI are facing a slew of lawsuits from artists who claim that their copyrighted material and personal information was scraped without consent or compensation. Ben Zhao, a professor at the University of Chicago, who led the team that created Nightshade, says the hope is that it will help tip the power balance back from AI companies towards artists, by creating a powerful deterrent against disrespecting artists’ copyright and intellectual property. Meta, Google, Stability AI, and OpenAI did not respond to MIT Technology Review’s request for comment on how they might respond. 

Zhao’s team also developed Glaze, a tool that allows artists to “mask” their own personal style to prevent it from being scraped by AI companies. It works in a similar way to Nightshade: by changing the pixels of images in subtle ways that are invisible to the human eye but manipulate machine-learning models to interpret the image as something different from what it actually shows. 

The team intends to integrate Nightshade into Glaze, and artists can choose whether they want to use the data-poisoning tool or not. The team is also making Nightshade open source, which would allow others to tinker with it and make their own versions. The more people use it and make their own versions of it, the more powerful the tool becomes, Zhao says. The data sets for large AI models can consist of billions of images, so the more poisoned images can be scraped into the model, the more damage the technique will cause. 

A targeted attack

Nightshade exploits a security vulnerability in generative AI models, one arising from the fact that they are trained on vast amounts of data—in this case, images that have been hoovered from the internet. Nightshade messes with those images. 

Artists who want to upload their work online but don’t want their images to be scraped by AI companies can upload them to Glaze and choose to mask it with an art style different from theirs. They can then also opt to use Nightshade. Once AI developers scrape the internet to get more data to tweak an existing AI model or build a new one, these poisoned samples make their way into the model’s data set and cause it to malfunction. 

Poisoned data samples can manipulate models into learning, for example, that images of hats are cakes, and images of handbags are toasters. The poisoned data is very difficult to remove, as it requires tech companies to painstakingly find and delete each corrupted sample. 

The researchers tested the attack on Stable Diffusion’s latest models and on an AI model they trained themselves from scratch. When they fed Stable Diffusion just 50 poisoned images of dogs and then prompted it to create images of dogs itself, the output started looking weird—creatures with too many limbs and cartoonish faces. With 300 poisoned samples, an attacker can manipulate Stable Diffusion to generate images of dogs to look like cats. 

Generative AI models are excellent at making connections between words, which helps the poison spread. Nightshade infects not only the word “dog” but all similar concepts, such as “puppy,” “husky,” and “wolf.” The poison attack also works on tangentially related images. For example, if the model scraped a poisoned image for the prompt “fantasy art,” the prompts “dragon” and “a castle in The Lord of the Rings” would similarly be manipulated into something else. 

Zhao admits there is a risk that people might abuse the data poisoning technique for malicious uses. However, he says attackers would need thousands of poisoned samples to inflict real damage on larger, more powerful models, as they are trained on billions of data samples. 

“We don’t yet know of robust defenses against these attacks. We haven’t yet seen poisoning attacks on modern [machine learning] models in the wild, but it could be just a matter of time,” says Vitaly Shmatikov, a professor at Cornell University who studies AI model security and was not involved in the research. “The time to work on defenses is now,” Shmatikov adds.

Gautam Kamath, an assistant professor at the University of Waterloo who researches data privacy and robustness in AI models and wasn’t involved in the study, says the work is “fantastic.” 

The research shows that vulnerabilities “don’t magically go away for these new models, and in fact only become more serious,” Kamath says. “This is especially true as these models become more powerful and people place more trust in them, since the stakes only rise over time.” 

A powerful deterrent

Junfeng Yang, a computer science professor at Columbia University, who has studied the security of deep-learning systems and wasn’t involved in the work, says Nightshade could have a big impact if it makes AI companies respect artists’ rights more—for example, by being more willing to pay out royalties.

AI companies that have developed generative text-to-image models, such as Stability AI and OpenAI, have offered to let artists opt out of having their images used to train future versions of the models. But artists say this is not enough. Eva Toorenent, an illustrator and artist who has used Glaze, says opt-out policies require artists to jump through hoops and still leave tech companies with all the power. 

Toorenent hopes Nightshade will change the status quo. 

“It is going to make [AI companies] think twice, because they have the possibility of destroying their entire model by taking our work without our consent,” she says. 

Autumn Beverly, another artist, says tools like Nightshade and Glaze have given her the confidence to post her work online again. She previously removed it from the internet after discovering it had been scraped without her consent into the popular LAION image database. 

“I’m just really grateful that we have a tool that can help return the power back to the artists for their own work,” she says.

#Fighting back against AI 'art' #Generative Art #Poisoning the Well

[milfglupshitto]

I keep seeing that nightshade post going around, here’s a direct link to the cited paper if anyone else wants to check it out (I’m going to try and read through it in full when my congestion clears up) but from my brief scan of intro + conclusion:

- the tool seems to be most effective not as a method to “corrupt” a batch of data but to alter the way a specific prompt is received by the code, effecting only that prompt (with a bit of splash to other prompts with higher levels of poisoning)

- this could allow an individual or company to establish a “do-not train filter” for their specific prompt (example used is Disney applying it to print images of “Cinderella”), which makes sense for artist styles (e.g. “do this in the style of Van Gogh”) but gets thorny with IP law as currently defined and enforced (who owns Cinderella? is this ownership fair or good for art?)

The paper’s conclusion paragraph, verbatim:

This work introduces the conceptual design, implementation and experimental evaluation of prompt-specific poison attacks on text-to-image generative image models. We believe our exploration of these issues shed light on fundamental limitations of these models. Moving forward, it is possible poison attacks may have potential value as tools to encourage model trainers and content owners to negotiate a path towards licensed procurement of training data for future models.

#ai art #machine learning #I think I’ve interpreted the doc language correctly but if anyone reads the paper and suggests a revision to my wording please add!

[corrupted-nightshade]

Burning Rage

Did I listen to Salvaged Rage for over 2 hours on loop? ...maybe.

{more info below to shorten post}

Since this is a vent art, i didn't really care much for making poses more accurate. only refs i used really was a picture of scraptrap and a pose ref only for the head position. ┐( ∵ )┌

___~___Info___~___

App used: Medibang

drawing tablet: XP-Pen

tools used: Sumi 4

Fandom: Five Nights at Freddy's

#my art #corrupted nightshade's art #art #digital art #digital drawing #five nights at freddy's #fnaf #william afton #the purple guy #2022 art #fnaf William #fnaf purple guy #scraptrap #fnaf scraptrap #fnaf afton #Henry emily #fnaf henry #fnaf henry Emily #fnaf 6 #five nights at freddy's 6 #fnaf pizza simulator #fnaf pizza sim #cw violence #cw choking

[poisonthefuckingwell]

"Going a step further to thwart AI copyright theft, the SAND Lab group has devised a prompt-specific poisoning attack targeting text-to-image generative models, which it has named Nightshade. “Nightshade poison samples are also optimized for potency and can corrupt a Stable Diffusion SDXL prompt in <100 poison samples,” write the researchers in their paper.

Whereas Glaze cloaks a single image – a process that can take hours on a laptop lacking a GPU – Nightshade operates on a much larger scale and could protect many more digital artworks.

Text-to-image poisoning could be achieved by simply mislabeling pictures of dogs as cats, so that when users prompted the model for a dog, the output would appear more cat-like. However, these rogue training data would be easy for AI models to reject in pre-screening. To get around this, the researchers curated a poisoned data set where anchor and poisoned images are very similar in feature space.

Feeding just 50 poisoned training samples into Stable Diffusion XL was sufficient to start producing changes in the generative AI text-to-image output. And by the time that 300 samples had been incorporated into the model, the effect was dramatic. A prompt for a hat produced a cake, and cubist artwork was rendered as anime."

It takes SO little nightshaded and glazed art to start dealing damage to their models. They are so fucking screwed

#anti ai #poisonthewell #important #glaze #midjourney

[thecreaturecodex]

Demon, Balor Lord, Ndulu

“Nefarox, Overlord of Grixus” © Wizards of the Coast, by Aleksi Briclot. Accessed at the GameLore Wiki here.

[Commissioned by @tar-baphon​, the second of four monsters revolving around Demon Lord Eltab. In the original Forgotten Realms lore, Ndulu was just the most powerful of the rivals to Eltab’s throne, but I wanted to build a little more frisson into their relationship (and explain how the Moaning Crown was broken without going into its thousand year backstory; see the Eltab entry for information on the Moaning Crown). Choosing to give Ndulu a polearm was solely due to this amazing artwork; I was uninspired by the various bits of balor art I could find, and looked slightly further afield.]

Demon, Balor Lord, Ndulu CR 23 CE Outsider (extraplanar) This mighty demonic figure is wreathed in flames and crowned with horns. It carries a massive polearm with a hooked blade.

Ndulu the Ghost Eater is a balor lord of great power and greater ambition. He is both servant to and enemy of Eltab, Lord of the Hidden Layer, who has coerced his service through layers of magical curses and mundane threats. Ndulu currently acts as a regent, occupying Eltab’s abandoned fortress and fighting other Abyssal powers that seek to rule his territory. Ndulu’s current plans involve finding a way to stop Eltab from making demoncysts and siphoning Abyssal power. If Ndulu succeeds, Eltab will be stripped of power and rank, becoming a nascent demon lord and therefore a much easier target for the Ghost Eater’s physical violence.

Unlike most balors, Ndulu disdains fighting with two weapons at once. He keeps a flaming whip as a back up weapon, but his primary armament is a massive polearm, which he wields with both power and finesse. His tactics involve dominating the battlefield with reach and mobility, and he can and has slain entire armies single-handedly. His epithet comes from his conquest of undead monsters; he can strike incorporeal creatures effortlessly, and can heal his wounds by slaying the living and undead both.

Ndulu’s forces include both powerful combatants such as vavakia and mariliths, and more subtle agents such as glabrezu and succubi. The latter he has sent into the camps of enemy demons and to work among mortals, both leads pursuing a way to cut off the power of the demoncysts. An adventuring party of good aligned characters might find themselves working for the Ghost Eater at several layers of remove, their mentors corrupted by a demon working for Ndulu to clean out demoncysts and sever their connection to the Abyss. Although referring to the creature as an ally may be a stretch, a nightwave nightshade keeps counsel with Ndulu, and elite units of nightwalkers and nightwings have fought for Ndulu’s cause.

Ndulu     CR 23 XP 820,000 CE Large outsider (chaotic, demon, evil, extraplanar) Init +13; Senses darkvision 60 ft., low-light vision, Perception +38, true seeing Aura flaming body, unholy aura (DC 27) Defense AC 40, touch 22, flat-footed 31 (+4 deflection, +9 Dex, +18 natural, –1 size) hp 492 (24d10+360) Fort +33, Ref +21, Will +29 DR 20/cold iron and good; Immune electricity, fire, poison; Resist acid 10, cold 10; SR 34 Defensive Abilities lifedrinker, negative energy affinity Offense Speed 40 ft., fly 90 ft. (good) Melee +2 vorpal unholy bill +39/+34/+29/+24 (2d6+23/19-20x3) plus 2d6 unholy) or +1 vorpal flaming  whip +38/+33/+28/+23 (1d4+15 plus 1d6 fire and entangle) or 2 slams +37 (1d10+14) Space 10 ft.; Reach 10 ft. (20 ft. with bill or whip) Spell-Like Abilities CL 20th, concentration +29 (+33 casting defensively) Constant—true seeing, unholy aura (DC 27) At will—dominate monster (DC 28), enervation, greater dispel magic, greater teleport (self plus 50 lbs. of objects only), haste, heroism, inflict critical wounds (DC 23), power word stun, telekinesis (DC 24), unholy blight (DC 23) 3/day—control undead (DC 26), horrid wilting (DC 27), quickened waves of fatigue 1/day—blasphemy (DC 26), fire storm (DC 27), summon (level 9, any 1 CR 19 or lower demon 100%), wail of the banshee (DC 28) Statistics Str 39, Dex 29, Con 40, Int 26, Wis 26, Cha 29 Base Atk +24; CMB +39; CMD 62 Feats Blind-fight, Combat Casting, Combat Reflexes, Following Step, Improved Initiative, Improved Critical (bill), Iron Will, Power Attack, Quicken Spell-Like Ability (waves of fatigue), Stand Still, Step Up, Step Up and Strike Skills Acrobatics +33 (+37 when jumping), Bluff +36, Diplomacy +36, Fly +38, Intimidate +36,  Knowledge (history) +32, Knowledge (nobility) +32, Knowledge (planes) +35, Knowledge (religion) +32, Linguistics +24, Perception +43, Sense Motive +35, Stealth +32, Use Magic Device +36; Racial Modifiers +8 Perception Languages Abyssal, Celestial, Common, Draconic, Necril, Protean, 13 others; telepathy 100 ft. SQ death throes, ghostly grasp, shorten grip, vorpal strike, whip mastery Ecology Environment any land or underground (Abyss) Organization unique Treasure standard (+2 unholy cold iron bill, +1 flaming whip, other treasure) Special Abilities Death Throes (Su) When killed, Ndulu explodes in a blinding flash of fire that deals 120 points of damage (half fire, half unholy damage) to anything within 100 feet (Reflex DC 37 halves). The save DC is Constitution-based. Entangle (Ex) If Ndulu strikes a Medium or smaller foe with its whip, he can immediately attempt a grapple check without provoking an attack of opportunity. If Ndulu wins the check, he draws the foe into an adjacent square. The foe gains the grappled condition, but Ndulu does not. Flaming Body (Su) A balor's body is covered in dancing flames. Anyone striking Ndulu with a natural weapon or unarmed strike takes 1d6 points of fire damage. A creature that grapples Ndulu or is grappled by him takes 6d6 points of fire damage each round the grapple persists. Ghostly Grasp (Su) Ndulu’s natural weapons and manufactured weapons deal full damage to incorporeal creatures as if they had the ghost touch weapon property. Lifedrinker (Su) Once per round when Ndulu kills a living or undead opponent, he can heal 150 hit points and remove any status effects as per a heal spell (CL 20th). Shorten Grip (Ex) Ndulu can shorten his grip on a polearm as an immediate action, threatening all squares within 10 feet instead of the polearm’s normal reach until he lengthens his grip with another immediate action. Vorpal Strike (Su) Any slashing weapon Ndulu wields gains the vorpal weapon quality. Weapons retain this quality for one hour after Ndulu releases the weapon, but after this the weapon reverts to its standard magical qualities, if any. Whip Mastery (Ex) Ndulu treats a whip as a light weapon for the purposes of two-weapon fighting, and can inflict lethal damage on a foe regardless of the foe's armor.

#ndulu #balor #demon #balor lord #pathfinder 1e #D&D #forgotten realms #commission #magic the gathering #mtg art

[wunderbud]

I've been on tumblr for basically a year now!! The art piece is gonna be a little late but genuinely I'd like to thank yall for an amazing year! I didn't think I'd get to over 50 followers let alone almost 400!! Like holy shit guys that's amazing!

Shout out to the friends I've made along the way, I genuinely wouldn't be here without all of you; @lockandkeyhyena @elesketchii @springbonniethewhore @aftonspogchamp @corpserabbit @misscloudiedays @lithiums-corner @3dfangs @abnormalceilingalt @bea-drawz-stuff @skelesani @artdumpandheadcanons @corrupted-nightshade and everyone else!

Yall are truly the best thing to ever happen to me! I wouldn't be anywhere without all of you! ❤️

[prisiidon]

https://www.tumblr.com/critter-of-habit/743064095987744769/what-is-this-about-the-tumblr-staff-wanting-to?source=share

if tumblr does end up partnering with m1djourney or other genAl (i get that tungl desperately needs funds), im gutted FR. nowhere is safe :')

Artist followers time2 use Nightshade (corrupts genAls) and/or Glaze (makes your style harder to be replicated) TTvTT. edit: one artwork may degrade poorly while another artwork may not, its hit or miss.

Last I heard the Editable Reblogs Xhit extension doesnt work anymore so hopefully there's an alternative, so that reblogs of our art are updated with the poisoned versions too.

#pri responds #i don't fully trust word of mouth but best be careful just incase this is legit

[pudgy-planets]

Unexpected friendship that actually makes a lot of sense when you think on it for a bit.

@readyplayerziggy

Sauce Brushing aside the fact they tower above majority of their peers, companions, and friends-

And have utterly humongous asses with the biggest difference being their busts/cup sizes-

Samantha is a soft-spoken, passionate, and empathetic woman who desires nothing more than to see people express their freedoms and embrace their heritage.

She despises corruption and authoritarianism, like that of the student council and their malefic Class President.

Which is why she "formed" the Apples Gang and eventually conjured her band of thirty misfits or so. They’re not delinquents in the traditional sense, rather they dress how they please, express their freedoms of protest and pull some hilarious pranks-

But they’re quite kind and welcoming despite the untruthful propaganda the council pushes against them. Samantha is kind until pushed. She has a lot going on deep down and rarely shows it.

Sana (in my canon at least) isn’t on the student council. She’s of course, leader of the Sumo Team with ChibiUsa, and president of the classical arts department. Not to be confused with Himiko’s role as the Cultural Arts and Drama department president.

However, she’s still a hardass and serious to an almost comical degree. Why? She has an artist’s passions and a true artistic soul. Therefore she finds beauty in all facets of life, not just paintings or traditional media, but the world itself.

The human body being the biggest form of artistic beauty she can imagine. So she finds Samantha utterly beautiful in that regard….

I mean wouldn’t you?

However, she recognizes, respects, and supports Samantha’s ambitions (as does Chikao). Heritage is beautiful and to repress its existence is a sin towards art an a means of expression.

So while their personalities differ, they see eye to eye on this front… also Samantha’s cats, Sabrina, Hex, Twitch, Alch, Nightshade, Merlin, Raven, the twins: Grim and Reaper, Banshee, Phantom, Shadow, Friday, Thirteen and Lady Charlie Higgleton Flobbenglagger Ploppingschmire Sinnimodana the First find Sana’s presence (and ass) unbelievably comfortable-

#muse: sana sunomiya #muse: samantha #readyplayerziggy #(speaking of which there’s this really good vore pic with sana- #well more of a sequence #planetary lore