#language processing in AI writing

[taslimursunybilas]

Can Artificial Intelligence Replace Human Writers? Unveiling the True Potential of ChatGPT and Google Bard

[frank-olivier]

Bayesian Active Exploration: A New Frontier in Artificial Intelligence

The field of artificial intelligence has seen tremendous growth and advancements in recent years, with various techniques and paradigms emerging to tackle complex problems in the field of machine learning, computer vision, and natural language processing. Two of these concepts that have attracted a lot of attention are active inference and Bayesian mechanics. Although both techniques have been researched separately, their synergy has the potential to revolutionize AI by creating more efficient, accurate, and effective systems.

Traditional machine learning algorithms rely on a passive approach, where the system receives data and updates its parameters without actively influencing the data collection process. However, this approach can have limitations, especially in complex and dynamic environments. Active interference, on the other hand, allows AI systems to take an active role in selecting the most informative data points or actions to collect more relevant information. In this way, active inference allows systems to adapt to changing environments, reducing the need for labeled data and improving the efficiency of learning and decision-making.

One of the first milestones in active inference was the development of the "query by committee" algorithm by Freund et al. in 1997. This algorithm used a committee of models to determine the most meaningful data points to capture, laying the foundation for future active learning techniques. Another important milestone was the introduction of "uncertainty sampling" by Lewis and Gale in 1994, which selected data points with the highest uncertainty or ambiguity to capture more information.

Bayesian mechanics, on the other hand, provides a probabilistic framework for reasoning and decision-making under uncertainty. By modeling complex systems using probability distributions, Bayesian mechanics enables AI systems to quantify uncertainty and ambiguity, thereby making more informed decisions when faced with incomplete or noisy data. Bayesian inference, the process of updating the prior distribution using new data, is a powerful tool for learning and decision-making.

One of the first milestones in Bayesian mechanics was the development of Bayes' theorem by Thomas Bayes in 1763. This theorem provided a mathematical framework for updating the probability of a hypothesis based on new evidence. Another important milestone was the introduction of Bayesian networks by Pearl in 1988, which provided a structured approach to modeling complex systems using probability distributions.

While active inference and Bayesian mechanics each have their strengths, combining them has the potential to create a new generation of AI systems that can actively collect informative data and update their probabilistic models to make more informed decisions. The combination of active inference and Bayesian mechanics has numerous applications in AI, including robotics, computer vision, and natural language processing. In robotics, for example, active inference can be used to actively explore the environment, collect more informative data, and improve navigation and decision-making. In computer vision, active inference can be used to actively select the most informative images or viewpoints, improving object recognition or scene understanding.

Timeline:

1763: Bayes' theorem

1988: Bayesian networks

1994: Uncertainty Sampling

1997: Query by Committee algorithm

2017: Deep Bayesian Active Learning

2019: Bayesian Active Exploration

2020: Active Bayesian Inference for Deep Learning

2020: Bayesian Active Learning for Computer Vision

The synergy of active inference and Bayesian mechanics is expected to play a crucial role in shaping the next generation of AI systems. Some possible future developments in this area include:

- Combining active inference and Bayesian mechanics with other AI techniques, such as reinforcement learning and transfer learning, to create more powerful and flexible AI systems.

- Applying the synergy of active inference and Bayesian mechanics to new areas, such as healthcare, finance, and education, to improve decision-making and outcomes.

- Developing new algorithms and techniques that integrate active inference and Bayesian mechanics, such as Bayesian active learning for deep learning and Bayesian active exploration for robotics.

Dr. Sanjeev Namjosh: The Hidden Math Behind All Living Systems - On Active Inference, the Free Energy Principle, and Bayesian Mechanics (Machine Learning Street Talk, October 2024)

Saturday, October 26, 2024

[sumarry]

Explore the differences between AI Writing and human-written content, how AI mimics human style, and what it means for the future of writing.

[jonhtv]

GravityWrite AI Tool: A Comprehensive Review and User Guide

GravityWrite AI Tool: A Comprehensive Review and User Guide In the fast-paced world of content creation, artificial intelligence tools are becoming essential for marketers, writers, and business owners. One of the most innovative and versatile AI writing tools to emerge is GravityWrite. GravityWrite AI Tool powered content creation platform promises to revolutionize how you generate, optimize,…

[neturbizenterprises]

Transform Your Content Creation with Deep Brain AI!

The Universe of technology is constantly expanding, and one of its newest stars is artificial intelligence (AI).

In this video, we explore how AI content creation, particularly through Deep Brain AI, is revolutionizing the way we produce engaging material.

Unlike traditional AI that relies on pre-programmed rules, generative AI creates new content from existing data. Deep Brain AI empowers us to overcome challenges like writer's block by generating ideas for blog posts, articles, and social media updates.

With advanced natural language processing and customizable templates, it streamlines our creative process while ensuring a consistent brand voice across platforms.

#DeepBrainAI #ContentCreation

[daveinediting]

It starts with a conversation about Shakespeare that triggers an old lesson about how humans only used spoken language for a long time before pictographs, hieroglyphics, and written language as we now know it. So in all that time you couldn't just write things down...

You had to remember them.

You had to accurately memorize them.

It turns out our ancestors memorized insane amounts of information through the spoken word. They had to develop that ability in order to pass acquired knowledge between communities and generations.

Memory was their only storage device. An organic storage device.

Once I got thinking on language... I remembered another lesson about the translation of languages and how sometimes one language maps multiple words onto one word in the target language. For example, eight words in ancient Greek onto the one English word, love. As in

I love my wife.

I love hamburgers.

Yeah. Awkward.

Another imperfect memory later and now we're being taught that the Lincoln-Douglas debates of 1858 lasted hours. Crowds gathered at these debates to listen, to engage with them also for hours. As in an hour-long opening statement by one candidate, an hour and a half-long response by the other candidate, and a half-hour rebuttal by the first candidate. The debates attracted crowds of up to 20,000 people including reporters and stenographers who covered the hours upon hours of debate.

Hours?

Yeah. Hours.

Woof.

I point these things out because they’re what made me wonder how different our historic predecessors must have been. After all, they could commit so much acquired knowledge to memory. Their brains were trained on the written word and the way in which the written word forms our understandings of the world. The resulting abilities ushered in deeper human understandings as well as sustained attention to the constructions of reasoned arguments.

I wonder how different these people might be who understood the world around them this way. I wonder how different their predecessors were whose tradition was spoken, whose knowledge was sustained and perpetuated through brute force memory.

How different were they, these people whose abilities are so far removed from our own?

I used to wonder if those abilities made our historical predecessors more capable than us in some way. After all, their oral and written traditions demanded much from them. Definitely their time. Definitely their mental bandwidth.

They exercised their intellects in ways we don't. Because we don't have to. The ways in which we now communicate and perpetuate knowledge bear lighter demands.

Which brings me back to Shakespeare.

Recently I heard a conversation with a professor challenging him to justify reading Shakespeare as a high school or college requirement when we can now understand Shakespeare through ChatGPT. We can generate fifty-word summaries and two hundred-word analyses of Shakespeare with AI and thus know and answer all there is about Shakespeare and his writing.

So why read him?

Seriously. Why?

That's just the tip of the argument, of course. Follow it all the way: Why should we be required to read anything? Novels. Short stories. Essays. What actual purpose does reading even serve when ChatGPT can boil it all down in seconds.

Is there a benefit of deeper knowledge on any subject whether it's a book, a short story, or an essay? And what do we get in exchange for our efforts to achieve such deeper understanding and knowledge. Does that effort, does that understanding, transform us in any objectively measurable way? And if not, does that understanding transform us in some perhaps more fundamental way.

Does it change us? Swap out our abilities like people who communicate primarily through 140 characters whose abilities replaced the abilities of people raised on radio then television whose abilities replaced the abilities of people raised on the written word whose abilities replaced the abilities of people raised on the spoken word.

What’s the actual prize for putting in the time and effort to read what someone else has committed to paper or screen? To deep dive into another human being's mind?

Because the oral tradition required it.

Because the written tradition demanded it.

And now?

Well? Is it or is it not simply good enough to just know what we need to know on demand?

Is access to knowledge the same thing as a deep understanding of that knowledge? And is there a difference that actually, you know, makes a difference?

Is the quality of our understanding really something to strive for anymore? Or is the tradition of study simply a mindless one that's made obsolete by knowledge on demand?

Ultimately, is there some advantage to a more muscular brain? One that’s gotta work harder, be more engaged in order to process the spoken and written word, on ideas and concepts and hypotheses and arguments on its way to understanding?

And.

Are we replacing that specific way of mental processing with something that makes our brains more muscular? More light weight? Or something in-between.

Is it that our mental abilities are now better tallied by the weight (such as it is) of our current mental musculature plus whatever exterior processes augment it like computers and smartphones and AI?

So we shouldn't sweat what we were formerly capable of and can't do now?

Is our resulting intellectual prowess, however it adds up, sufficient for successfully and sustainably navigating our stormy Present that’s seized in a constant state of rapid and relentlessly whirling transformation?

Or is it essentially a product of that change.

And.

Are we fine-tuned for this age of human existence…

Or are we not.

😕

[m-accost]

PSA: If you see a screenshot of the basic GPT-3.5 version of ChatGPT producing poetry that is remotely novel or clever or actually a decent pastiche of a given poet, it's probably faked. It's been RLHFed to output, at the merest use of the words "poem" or "poetry", doggerel of the sort that tends to get posted on restaurant walls or sent to local newspapers. Simply telling it "poem" or "poetry" will prompt it to produce a bit of doggerel about poetry:

[creativestudio391-blog]

GoogleMUM #algorithm #search #seo #update #seoai

https://www.seolady.co.uk/google-mum-update-ai-ranking-seo-nlp-multitask-unified-model/ What is Google MUM and when was it updated in 2023?

content #naturallanguageprocessing #AI #Ranking

Multitask Unified Model MUM: Google developed its Natural Language Processing NLP - Does This Mean A New Way for SEO?

★ Originally dubbed "BackRub", the baby Google was the result of a research project that started back in 1996. Which means in 2026, technically, Google will be 30 years old. Do you remember Ask Jeeves, Netscape and Yahoo chat rooms? MSN chat and My Space with Tom?

In 2023, MUM uses natural language processing (NLP) and deep learning techniques to interpret complex questions and return highly relevant results. Google has famously been tight lipped with most new releases, from their blog the first announcement was in 2021.

★ It was created to address the growing need for more advanced search capabilities and to provide better experiences for users seeking information. ChatGPT AI was generally released in November 2022, and the SEO circles around the world are keeping MUM at the forefront of their curiosity in 2023.

★ https://www.seolady.co.uk/seo-keyword-research-chatgpt-search-phrases-long-tail-synonyms/

The MUM update is Google’s new AI language model that uses natural language processing to improve search results. This model allows Google to understand more complex queries, and it can help provide more accurate and relevant search results. MUM can also understand and translate between multiple languages, making it a powerful tool for international search.

★ According to Google, the MUM update is 1000 times more powerful than its previous BERT update, which was already a significant improvement in natural language processing. With MUM, Google can understand longer and more complex queries, making it easier for users to find what they’re looking for. This update also brings new features like a new search experience with dynamic layouts and visual search, so users can find the information they need faster and more efficiently.

website #multitaskunifiedmodel #bert #nlp #searchqueries #machinelearning

[laiqverse]

Mastering the Art of Communicating with AI

As we enter a new era of technological advancement, communication with artificial intelligence (AI) has become increasingly important. The art of communicating with AI is an essential skill that individuals and businesses alike must possess to navigate this new terrain. To start, it is essential to understand the intricacies of AI language. Unlike human language, which can be ambiguous and…

[frank-olivier]

Rethinking Content, Reimagining AI: The Forward-Thinking Philosophy of Notebook LM

Google's Deep Mind project, Notebook LM, represents a pivotal moment in the intersection of Artificial Intelligence (AI) and content creation. By generating insightful analyses, explanations, and podcasts from diverse uploaded sources, Notebook LM revolutionizes the way individuals interact with and process information. This innovative tool, born out of the experimental "Project Tailwind" within Google Labs, epitomizes the quest for harnessing language models to enhance learning and idea organization.

The far-reaching applications of Notebook LM are a testament to its versatility. Professionals, including writers, journalists, and sales teams, leverage the platform to streamline research and facilitate knowledge sharing, thereby augmenting productivity. Conversely, personal use cases, such as uploading journals for introspective analysis, demonstrate Notebook LM's profound potential in fostering self-awareness and personal growth. This dual utility underscores the democratizing effect of Notebook LM, empowering the creation of niche content that may have been economically unviable through traditional podcasting channels.

The "Audio Overview" feature, with its remarkably human-like conversational style, significantly enhances user engagement. However, this very aspect also precipitates crucial inquiries into the dynamics of trust and perception in AI-generated content. As the creative outputs of humans and AI become increasingly indistinguishable, the boundaries between them begin to blur, prompting a necessary reevaluation of AI's role in creative processes. This convergence necessitates a thoughtful, user-centric approach to design and education, ensuring that the benefits of AI-driven content are maximized without diminishing the inherent value of human creativity.

The development of Notebook LM is also distinguished by a laudable emphasis on responsible AI innovation. The integration of robust safety filters, stringent privacy measures – including the ephemeral "context window" for user uploads – and a proactive stance against potential misuse, collectively demonstrate a forward-thinking commitment to mitigating the risks associated with AI. This paradigmatic approach is indispensable in the rapidly evolving AI landscape, where the interplay between innovation and responsibility is continually being redefined.

As we look to the future of content creation, Notebook LM shines as a pioneering example of what’s possible when human creativity and AI seamlessly intersect. This powerful fusion has the potential to revolutionize industries far and wide. However, as we tap into its transformative power, we’re also faced with complex questions around AI-driven creativity, trust, and innovation. As we embark on this exciting technological journey with Notebook LM, it’s crucial that we thoughtfully consider its broader implications, working to ensure that the benefits of AI are shared by all and its challenges are met with careful consideration

Raiza Martin & Steven Johnson: Inside NotebookLM (Google DeepMind, November 2024)

Wednesday, November 27, 2024

[reasonsforhope]

"The Writers Guild has reached a tentative agreement with the Alliance of Motion Picture and Television Producers to end its strike after nearly five months. The parties finalized the framework of the deal Sunday when they were able to untangle their stalemate over AI and writing room staffing levels.

“We have reached a tentative agreement on a new 2023 MBA, which is to say an agreement in principle on all deal points, subject to drafting final contract language,” the guild told members this evening in a release, which came just after sunset and the start of the Yom Kippur holiday that many had seen deadline to wrap up deal after five days of long negotiations...

Despite today’s welcome news, it still will take a few days for the strike to be officially over as the WGA West and WGA East proceed with their ratification process. During the WGA’s last strike in 2007-08, a tentative agreement was reached on the 96th day and it wasn’t over until the 100th...

All attention will now turn to ratifying the WGA deal and getting SAG-AFTRA and the AMPTP back to the bargaining table to work out a deal to end the actors’ strike, which has now been going on for 70 days.

Details of the WGA’s tentative agreement haven’t been released yet but will be revealed by the guild in advance of the membership ratification votes. Pay raises and streaming residuals have been key issues for the guild, along with AI and writers room staffing levels."

-via Deadline, September 24, 2023

[notherpuppet]

What is your opinion of AI? Personally, I think that like any technology, it depends on the user and their intentions, but that is just me.

What about you?

1. Theft

The most central issues with AI as it is now is that the programs were trained/are trained with STOLEN art. Stolen visual art, music, writing, etc.

The vast majority of what it has been fed is stolen. As in, the artists behind the work were not ever given the chance to consent nor be compensated for their works being used to feed the machine.

This reason alone is straight up copyright infringement and the optimist in me does believe the long arm of the law is gonna shut these programs down for that. But the long arm of the law is looooooong, and the technology is disrupting people’s livelihoods now. Unlike robots or machinery that was invented and built to expedite assembly line/factory work, this technology is only functional by using other people’s labor. If we didn’t live in a society where you have to “earn” your right to live in it, then this would still be wrong, but it probably wouldn’t be such an existential problem.

There are active class action lawsuits for infringement of copyright. And the private sector has begun filing suits and I’m quite certain they’ll win because again—it’s simply theft. These companies did not make licensing contracts, they’re not paying royalties to the artists they stole from.

So if you consider using ai that generates “art” (whether it is visual, music, writing, etc.) please consider stopping immediately, as you would actively be benefiting from theft (which is wrong imo!!!!)

2. AI in its present form dishonors the human spirit

In my personal relationship with AI technology, I do not use it to generate ideas or ‘art’. I detest the notion to use technology in that way tbh. AI is a form of technology, so it’s difficult to break it down into every specific use it actually has. But here’s an attempt; no to generative AI, okay to certain AI.

There are kinds of AI programming in the programs I use (such as features that help you color in a shape quickly or make a perfect circle). This is useful tech (that requires zero IP theft) and I like it because it helps me by taking care of tedious tasks so that I have more time to spend in the creative and drawing processes. But I still choose the colors, I still draw the images, I still write the stories.

I think the way AI is used right now with a focus on “creative thinking” (where it’s not actually creating anything it’s just churning out other people’s *stolen* ideas and practice) is a total waste. AI being used as an assistant to help humans find information easily can be/has been swell. And requires no theft :D

But for whatever reason (greed, capitalism are my guesses), tech companies are leaning into a direction to replace creativity with AI?? I imagine the people behind this view the practice of art as tedious work because it is challenging??

But the beauty of art and the practice of it is that it allows humans to experience and overcome challenges with little to no stakes.

When society determines that is not a valuable use of human time, then I think we’ll all be significantly more miserable. If we allow a machine to be “creative” and leave us to only experience challenges with stakes—like survival (rent, putting food on the table).

So here are some examples of how I feel about AI uses;

AI to translate languages, find resources, discern malicious malware/spam from harmless messages > 👍🏽

AI to generate ideas/art for you > 🤢 Why??????? Why would you want that…that’s the most exceptional part of the human experience and you relinquish it to a bot trained on stolen ideas? 😭

[nowoyas]

Trying to make sense of the Nanowrimo statement to the best of my abilities and fuck, man. It's hard.

It's hard because it seems to me that, first and foremost, the organization itself has forgotten the fucking point.

Nanowrimo was never about the words themselves. It was never about having fifty thousand marketable words to sell to publishing companies and then to the masses. It was a challenge, and it was hard, and it is hard, and it's supposed to be. The point is that it's hard. It's hard to sit down and carve out time and create a world and create characters and turn these things into a coherent plot with themes and emotional impact and an ending that's satisfying. It's hard to go back and make changes and edit those into something likable, something that feels worth reading. It's hard to find a beautifully-written scene in your document and have to make the decision that it's beautiful but it doesn't work in the broader context. It's fucking hard.

Writing and editing are skills. You build them and you hone them. Writing the way the challenge initially encouraged--don't listen to that voice in your head that's nitpicking every word on the page, put off the criticism for a later date, for now just let go and get your thoughts out--is even a different skill from writing in general. Some people don't particularly care about refining that skill to some end goal or another, and simply want to play. Some people sit down and try to improve and improve and improve because that is meaningful to them. Some are in a weird in-between where they don't really know what they want, and some have always liked the idea of writing and wanted a place to start. The challenge was a good place for this--sit down, put your butt in a chair, open a blank document, and by the end of the month, try to put fifty thousand words in that document.

How does it make you feel to try? Your wrists ache and you don't feel like any of the words were any good, but didn't you learn something about the process? Re-reading it, don't you think it sounds better if you swap these two sentences, if you replace this word, if you take out this comma? Maybe you didn't hit 50k words. Maybe you only wrote 10k. But isn't it cool, that you wrote ten thousand words? Doesn't it feel nice that you did something? We can try again. We can keep getting better, or just throwing ourselves into it for fun or whatever, and we can do it again and again.

I guess I don't completely know where I'm going with this post. If you've followed me or many tumblr users for any amount of time, you've probably already heard a thousand times about how generative AI hurts the environment so many of us have been so desperately trying to save, about how generative AI is again and again used to exploit big authors, little authors, up-and-coming authors, first time authors, people posting on Ao3 as a hobby, people self-publishing e-books on Amazon, traditionally published authors, and everyone in between. You've probably seen the statements from developers of these "tools", things like how being required to obtain permission for everything in the database used to train the language model would destroy the tool entirely. You've seen posts about new AI tools scraping Ao3 so they can make money off someone else's hobby and putting the legality of the site itself at risk. For an organization that used to dedicate itself to making writing more accessible for people and for creating a community of writers, Nanowrimo has spent the past several years systematically cracking that community to bits, and now, it's made an official statement claiming that the exploitation of writers in its community is okay, because otherwise, someone might find it too hard to complete a challenge that's meant to be hard to begin with.

I couldn't thank Nanowrimo enough for what it did for me when I started out. I don't know how to find community in the same way. But you can bet that I've deleted my account, and I'll be finding my own path forward without it. Thanks for the fucking memories, I guess.

[i-will-physically-fight-you]

I think something that's fascinating in the AI discussion is how non-creatives perceive AI versus how many creatives perceive AI.

For example, years before AI was a thing--I spoke with someone about my creative writing projects and they expressed to me how they found it unfathomable that I could just make up entire worlds far removed from our reality of existence. To them, it was like magic.

To me, it was the culmination of countless hours spent playing with words until they flowed into semi-coherent lines of thought and emotion. I remember being ten years old and laboring away on my "biggest" novel project ever--it was 5k words full of singular sentence-long paragraphs and garbled heaps of grammar atrocities to the English language.

If I hadn't written it, I wouldn't have come to learn how to create the basic foundations of a story.

But I do get the "it's magic" sentiment a bit--I'm that way with music. Theoretically, I understand the components of a music composition but it feels like magic to see a musician that can listen to a tune for the first time and play it perfectly due to years of honing in their craft.

That's the premise of that quote from Arthur C. Clarke: "Magic's just science we don't understand yet."

When it comes to anything we don't have countless hours of experience with, it feels like magic. It feels like something that's outside of our feeble human capabilities. It's not until we start to put in the time to learn a skill that it becomes more attainable inside our heads.

Generative AI presents a proposition to the non-creative: "What if you could skip past the 'learning process' and immediately create whatever art of your choosing?"

It's instant dopamine. In a world that preys upon our ever-decreasing attention spans and ways of farming short spikes of dopamine, was it ever a surprise that generative ai would be capitalized in this fashion?

So for the non-creative, when they use generative AI and see something resembling their prompt, it feels good. They are "writing" stories, they are "making" art in ways they could never do with their lack of skills.

(It is, in fact, really cool that we have technology that can do this. It's just incredibly shitty that it's exploitative of the human artists whose works were taken without permission as well as its existence threatening their livelihoods.)

What I think is equally concerning as the data scraping of generative ai is the threat that AI imposes on the education of the arts. More and more, you see an idea being pushed that you don't need knowledge/experience in how to create art, all you need to do is feed prompts into generative ai and let it do the "work" for you.

Generative AI pushes the idea that all art should be pristine, sleek and ready for capitalism consumption. There is no room for amateur artists struggling like foals to take their first steps in their creative journeys. We live in a world where time is money and why "waste" time learning when you can have instant success?

It's a dangerous concept because presents a potential loss in true understanding of how art works. It obscures it and makes it seem "impossible" to the average person, when art is one of the freest forms of expressions out there.

It's already happening--Nanowrimo, the writing challenge where the entire point was writing 50k original words in a single month regardless of how pretty it looked--coming out and saying that it is ableist and classist to be opposed to AI is the canary in the coalmine of what's to come.

For the non-creatives who enjoy the generative ai, it feels like a power fantasy come to life. But for creatives concerned about generative ai?

We're living in a horror movie.

[neturbizenterprises]

Transform Your Content Creation with Deep Brain AI!

The Universe of technology is constantly expanding, and one of its newest stars is artificial intelligence (AI).

In this video, we explore how AI content creation, particularly through Deep Brain AI, is revolutionizing how we produce engaging material.

Unlike traditional AI which relies on pre-programmed rules, generative AI creates new content from existing data. Deep Brain AI empowers us to overcome challenges like writer's block by generating ideas for blog posts, articles, and social media updates. With advanced natural language processing and customizable templates, it streamlines our creative process while ensuring a consistent brand voice across platforms.

[mikaela-the-slut-expert]

Hiii! I have a request.

Something that starts with smut and ends with angst.

My idea is that Hualian and we would be having casual sex, until one had the idea to try to do something rougher. Of course, before doing so, they asked for our consent, and believing that it would be no big deal, we accepted.However, after the first slap we receive, s/o ends up being a little shocked and scared, but not wanting to spoil the fun he kind of accepts.The smut would continue normally, our character would break even when he was in a position that was kind of "suffocated" (it can be the one that the person takes in the ass and in the mouth at the same time, wow it was funny to write this 😂), and because he can't breathe properly and already feeling pain due to previous actions, the character starts to cry and pats the thigh of whoever was in front of him. As soon as his mouth is freed, he speaks the safe word while his body trembles in terror.I would like it to end with anguish or comfort, but I leave that to the writer's discretion!

The idea in general is this, but if you feel that an ending with only dialogue doesn't look so cool, you can include a soft smut!!Thank you for your attention and please take care of yourself ;) Also, sorry if there were any grammatical mistakes, English is not my first language.

Too Much

Hualian x gn!reader

Tw!!! Safe word usage, NSFW, slapping

I just don't feel like my writing has been up to par lately y'all 🖐️😃🔫

____________________________________

You're sandwiched

Literally.

It's one of those nights when all three of you happen to be free of duties, and all of you are home. So what do you guys do when you're all home? Have sex.

It's nothing over the top, just casual, cuddly, sleepy, gentle sex. Hua Cheng is slowly fucking you from behind, but just because he's slow doesn't mean he isn't reaching deep. His dick is big and Hua Cheng knows how to use it, effectively hitting that spot that makes you moan and cry for more.

Xie Lian sits on the edge of the bed near your head. He runs his fingers through your hair and every so often he'll share kisses with you or San Lang. No one knows why and nobody asks, he just has a thing for watching.

You're lying on your back, eyes shut, with soft whimpers at Hua Cheng takes his time with you. Xie Lian runs the back of his nails across your cheek to get your attention. "Baobei, do you think you would be alright if this one were to get a little rough with you?" He hums softly smiling down at you.

Xie Lian always asks, he never dares to put his hands on you without permission. San Lang stares and waits for your answer, because although he won't be hitting you he'd like to fuck you deeper if you'd allow him to. And of course you trust your lovers wholeheartedly, you've gotten rough before so what's new? "Yes, that's fine"

As soon as yes slips past your lips Hua Cheng takes the invitation to quicken the smack of his hips against your ass. He loves to fuck you fast and deep, he really just likes to make you feel good. You yelp when he suddenly flips you onto your stomach and plunges into your hole again. Hua Cheng easily manhandles you around, bruising clutch on your hips and thighs.

Xie Lian catches your attention when he moves in front of you so that your chin rests on his soft, milky thigh. You go to say something to him but you're interrupted when his hand slaps across your cheek. "Don't speak unless you're spoken too, Qin Ai De"

It was relatively normal. It wasn't unusual for Xie Lian to get a little rough with you and make rules for you. It wasn't uncommon for him to say degrading things if you had allowed him to. The slap hurt and you're barely able to process it with how rough Hua Cheng is fucking you. It honestly leaves you a little frightened and on edge. You don't know why, this is a relatively normal bedroom activity but today it doesn't sit right with you.

Maybe you're just overthinking it, so you let them throw you around some more and you let the night keep going. It's not until later into the night that everything goes horribly wrong.

Hua Cheng is still fucking you from behind but Xie Lian has no joined in on the fun and he's fucking your throat. Also not uncommon. It was usual for you to let them do this, keep all your holes filled at all times. You enjoyed it usually but today it feels awful and no matter how much they fuck into you or talk to you, you don't feel good. Your skin pricks with discomfort and you can't stop squirming which had earned you a few more slaps.

You haven't said anything though, you wanted to make your lovers feel good they deserved that much but you can't. You feel like you're going to hurl on Xie Lian's dick if they don't stop. You feel disgusting, and smothered. It feels suffocating.

Tears align your vision and your hand smacks against Xie Lian's thigh. Albeit confused, he immediately pulls back and Hua Cheng pulls out as well. You cough and spit up mumbling the safe word through harsh pants and that's all it takes for them to know tonight's done.

Hua Cheng gets off the bed and you don't really get an opportunity to see where he went because Xie Lian holds your face in his hands. "What's wrong? Did we go too far, baobei?"

You want to answer but you end up just sobbing in his hands. He lets you curl up on his lap while he threads his fingers through your hair, cooing at you and telling you that you had been very good, that you had done well. Hua Cheng comes back with warm towels to clean the three of you up, and he carefully washes whatever swear or stickiness covers your body.

It's silent for a while, not an uncomfortable silence, calm and safe. You eventually speak up. "I'm sorry... I- when you slapped me I got scared and it just set me off for the whole night. I don't know why, I'm sor-"

Hua Cheng kisses the corner of your mouth to shush you. "Sorry for what? You do not need a reason to feel unsafe or uncomfortable. We'll always stop if you are. Do not feel sorry for us" Hua Cheng lies atop of you like a log. His face is scrunched, like he's personally offended at himself for not noticing.

Xie Lian runs his fingers across your cheek, peppering soft kisses across your face. "It's alright, this one apologizes. I won't slap you in the future, Qin Ai De. You know we love you dearly, we would never hurt you purposefully. You must tell us immediately next time Y/n we don't wish to make you feel this way ever again." his brows are furrowed and you can tell he's probably beating himself up over it.

You don't feel suffocated anymore. You nuzzle Xie Lian's leg. Mumbling a soft "Okay... M'love you both"

They both know that there won't be anymore sex tonight so they both cuddle up next to you and dote on you until you fall asleep.