#chatgpt clone development | Explore Tumblr posts and blogs

jcmarchi · 3 months ago

Text

AI and Human Collaboration in Creating Short Series: How to Find Balance and Shape the Industry

New Post has been published on https://thedigitalinsider.com/ai-and-human-collaboration-in-creating-short-series-how-to-find-balance-and-shape-the-industry/

AI and Human Collaboration in Creating Short Series: How to Find Balance and Shape the Industry

Days of multi-million dollar budgets to make a movie or TV series are over. Now you can add special effects to an episode for $100 instead of $8,000. Trust me, no one can tell the difference. Creating exciting dialogues and plot twists takes weeks instead of months. Voiceover and dubbing in dozens of languages are done in a couple of clicks. Yes, AI is already heavily involved in the film industry, and this is just the first step.

It is expected that by 2033, Global AI in Film Market size will be around 14.1 Billion, with a CAGR of 25.7%. TV series will be released faster than before –– with AI, editing time is reduced by 40%. Certainly, more content will affect profits. It is projected to grow between $80 billion and $130 billion annually in the media and entertainment industry.

What will happen to actors, writers, producers, and others? Will they be left behind? Only if they fail to adapt. Human-AI synergy is a key. And here’s how to achieve it at all stages of content production.

Human-AI Brainstorm in Screenwriting

You all use ChatGPT (or other chatbots) to write emails and social media posts, so why shouldn’t screenwriters use AI to create cool stories? That’s exactly what 29% of the industry members are already doing. But here’s the thing: just like AI-generated text isn’t perfect, writing a script with AI alone would be boring.

This is where the human creative brain is needed. It’s the one who can guide the AI, fill it with really interesting ideas, and turn the text into a TV show. Hooks, dialogues that you can’t tear yourself away from, sharp plot twists –– it’s all a creative brainstorm of both AI assistants and humans.

Actors Enhancing, not Replacing

AI can be used to “resurrect” actors and bring movie favorites back to the screen. For instance, Grand Moff Tarkin in “Rogue One: A Star Wars Story” is a digital clone of Peter Cushing after his death. James Dean, who died back in 1955 in a car accident, will soon reappear on screen in “Back to Eden”. Using the same technology, Carrie Fisher appeared as Princess Leia in the “Star Wars” movie.

Another way AI is interacting with the movie industry is by transforming age representation. AI can easily make actors older or younger, if needed. This is exactly what Disney did with Harrison Ford in “Indiana Jones and the Dial of Destiny. Based on her persona from the previous movies, they created a new, younger version.

And finally, face changing. A very handy and extremely popular thing in the near future. Let’s say a team of actors worked together to create a series, and producers want to come up with something new with the same cast. Multiple products with the same actors wouldn’t work. In that case, Faceswap allows replacing actors’ faces �� and now it’s completely different characters.

What about human actors? They can profit from this! Actors can sign an agreement to use their faces in different shows and earn passive income. Besides, in this way, they can increase their recognition and then ask for higher rates.

Production Timeline and Budgets are Cut

It takes months to a year to shoot a season of a traditional TV series. Don’t even talk about enormous budgets! The thing about short TV series is that they cost about $200-300k –– that’s about 10 times cheaper than the classic format. And user engagement is higher! AI helps a lot with that.

Make special effects? 42% of industry professionals already use AI to do so. Fly halfway around the world to shoot an episode in the tropics? No more needed, as you can generate the background with AI!

And while AI goes further and further in its enhancements, don’t ignore the irreplaceable part played by humans. They are responsible for story direction, creative vision, making sure the audience gets the picture and the action on the screen, and much more!

Post-Production Takes Weeks, Not Months

While post-production was once the most time-consuming part of working on a TV series, everything is changing now. AI is reshaping the way video editing works –– what used to take months is now a matter of weeks.

The VFX tool makes creating animation effects a breeze, and it doesn’t require a huge team anymore. The same goes for dubbing and localization –– with AI, you can quickly transform content for a global audience. For example, with ElenLabs’ dubbing tool, actors’ voices are changed within a few clicks, translated into different languages, while retaining their intended emotionality. Keeping scenes authentic no longer requires a whole studio.

AI has also sped up and facilitated dozens of editing team members. Color correction, sound effects, visual cleanup –– all these processes are highly streamlined, so you can ramp up content production.

Interactions with Actors

The latest technologies let viewers interact with characters while watching their favourite shows. How about chatting with them like friends? Tools like character.AI make it possible – definitely a game changer! Surely, this kind of interaction has clear rules to keep the actors safe. No intimate texting, and all communication must be related to the series.

The human actor’s participation is not needed. He’s got a digital twin who’s trained on his behaviour.

This interaction benefits actors –– they can license their likenesses and personalities for AI-driven interactive experiences. Something like musicians get royalties for their hits. The amount depends on the time spent chatting with the audience and its volume.

What’s Next?

In the near future, viewers will be even more involved in the characters’ lives and series! They gradually move from passive observers to participants in its creation.

How about inventing a personalized series?

With AI, anyone could make a rom-com starring Johnny Depp and Leonardo DiCaprio or add their own Roman Empire atmosphere! It’s called setting. And it will happen even faster than you think –– there are already tools that allow it, so it’s a matter of the next couple of years.

Another trend that is definitely going to grow is vertical short series. They are produced faster, engage much more efficiently and, most of all, are adapted to the modern viewer’s needs, who find it hard to keep their attention for an hour-long series.

We are now seeing a new round of development of the AI-enabled TV series industry. The Human-AI balance is inevitable here. AI is helping screenwriters create compelling stories, actors are using AI to make money, directors are bringing bold ideas to life with tight budgets, and post-production teams are working faster.

So, pack your popcorn and get ready to be amazed!

0 notes

miracuvesseo · 5 months ago

Text

ChatGPT Clone by Miracuves: Revolutionizing Conversational AI

The digital age is defined by innovation, and one of the most transformative technologies to emerge in recent years is conversational AI. Among the leaders in this field is OpenAI’s ChatGPT, a powerful tool for automating interactions, enhancing customer experiences, and streamlining business processes. However, as businesses strive for more tailored solutions, Miracuves has stepped up to the plate, offering a ChatGPT Clone that redefines the capabilities of conversational AI.

If you’re looking to build a chatbot tailored to your unique needs, Miracuves’ ChatGPT Clone is the perfect solution. Let’s explore how it works, its features, and why it’s the future of AI-driven communication.

What is the ChatGPT Clone by Miracuves?

Miracuves’ ChatGPT Clone is a customizable conversational AI platform inspired by OpenAI’s groundbreaking ChatGPT model. It provides businesses with a ready-to-deploy solution for creating chatbots and virtual assistants that cater to specific industries, functions, and user requirements. Unlike off-the-shelf AI tools, this clone is designed for scalability, personalization, and seamless integration with various business systems.

The ChatGPT Clone from Miracuves is not just a replica; it’s an enhanced version offering flexibility and control, ensuring businesses can align the AI’s capabilities with their unique goals.

Key Features of Miracuves’ ChatGPT Clone

Customizable Personality The clone can be tailored to reflect your brand’s tone and style. Whether you want a formal assistant for professional settings or a friendly chatbot for customer support, the personality is fully customizable.

Multilingual Capabilities Break down language barriers with multilingual support. Miracuves’ ChatGPT Clone enables businesses to engage with global audiences effortlessly.

Advanced NLP and Context Awareness The AI leverages cutting-edge Natural Language Processing (NLP) to understand and respond to user queries effectively. It retains context within conversations, making interactions seamless and human-like.

Industry-Specific Training Miracuves provides the ability to train the ChatGPT Clone on industry-specific data, ensuring accurate and relevant responses for specialized use cases.

Omnichannel Support Deploy the AI across multiple channels, including websites, mobile apps, messaging platforms, and more, ensuring consistent and accessible communication.

Data Security and Privacy Host the clone on private servers or integrate it with your existing systems to maintain full control over sensitive data. Miracuves adheres to strict data security protocols to safeguard information.

Seamless Integration The ChatGPT Clone can connect with CRMs, ERPs, and other enterprise tools, enabling a unified and efficient workflow.

Real-Time Analytics Track performance and user interactions with robust analytics tools. Gain insights to continuously refine the chatbot’s effectiveness.

Applications of Miracuves’ ChatGPT Clone

Customer Support Automate 24/7 customer service with intelligent responses to FAQs, order tracking, and troubleshooting, reducing workload and response times.

E-Commerce Assistance Improve online shopping experiences by guiding customers, providing product recommendations, and assisting with purchases.

Education Create virtual tutors that answer student questions, explain concepts, and deliver personalized learning support.

Healthcare Develop AI assistants to answer health-related queries, book appointments, and share information about medical services.

Content Creation Generate engaging content such as blogs, social media posts, and marketing copy using the AI’s language generation capabilities.

Finance and Banking Automate customer interactions in banking, provide financial advice, and guide users through complex procedures.

Recruitment and HR Streamline hiring processes with AI-powered chatbots that screen candidates, answer applicant queries, and manage scheduling.

Why Choose Miracuves’ ChatGPT Clone?

Tailored Solutions Unlike generic AI models, Miracuves offers full customization, ensuring the solution aligns perfectly with your business objectives.

Cost-Effective Save on subscription costs by deploying your own conversational AI model. With Miracuves, you pay for what you need, making it an affordable long-term investment.

Faster Time-to-Market With Miracuves’ ready-to-deploy ChatGPT Clone, businesses can launch their AI-powered chatbot in record time.

Full Control and Ownership Maintain control over the AI’s training data, deployment, and functionalities, ensuring compliance with regulations and internal standards.

Scalability Whether you’re a startup or an enterprise, the ChatGPT Clone is designed to grow with your business, handling increased traffic and complex requirements with ease.

How to Get Started with Miracuves’ ChatGPT Clone

Define Your Goals Start by identifying the purpose of your chatbot. Is it customer service, lead generation, or personalized support? Understanding your goals will guide the customization process.

Collaborate with Miracuves Work with Miracuves to design and deploy your ChatGPT Clone. Their team will ensure the AI is tailored to meet your specific needs.

Train the AI Provide domain-specific data to train the AI for optimal performance. Miracuves’ expertise ensures a smooth and effective training process.

Deploy and Optimize Launch the chatbot on your desired platforms and refine it based on user feedback and analytics.

Future of ChatGPT Clones

As conversational AI evolves, clones like Miracuves’ ChatGPT are becoming indispensable tools for businesses. The future holds even more possibilities, including emotional intelligence in bots, integration with AR/VR environments, and hyper-personalized user interactions. By adopting these technologies early, businesses can stay ahead of the curve and deliver unparalleled value to their users.

Conclusion

Miracuves’ ChatGPT Clone is more than just an AI chatbot—it’s a gateway to smarter, more efficient, and highly personalized communication. Whether you’re looking to improve customer support, streamline operations, or enhance user engagement, this clone provides the tools and flexibility to make it happen.

In a world where conversational AI is reshaping industries, Miracuves stands out by offering a solution that combines innovation, customization, and practicality. If you’re ready to revolutionize the way you interact with your audience, it’s time to explore the possibilities of the ChatGPT Clone by Miracuves.

Empower your business with Miracuves – the future of conversational AI.

#app development #ChatGPT Clone

0 notes

mariacallous · 1 year ago

Text

Last week OpenAI revealed a new conversational interface for ChatGPT with an expressive, synthetic voice strikingly similar to that of the AI assistant played by Scarlett Johansson in the sci-fi movie Her—only to suddenly disable the new voice over the weekend.

On Monday, Johansson issued a statement claiming to have forced that reversal, after her lawyers demanded OpenAI clarify how the new voice was created.

Johansson’s statement, relayed to WIRED by her publicist, claims that OpenAI CEO Sam Altman asked her last September to provide ChatGPT’s new voice but that she declined. She describes being astounded to see the company demo a new voice for ChatGPT last week that sounded like her anyway.

“When I heard the release demo I was shocked, angered, and in disbelief that Mr. Altman would pursue a voice that sounded so eerily similar to mine that my closest friends and news outlets could not tell the difference,” the statement reads. It notes that Altman appeared to encourage the world to connect the demo with Johansson’s performance by tweeting out “her,” in reference to the movie, on May 13.

Johansson’s statement says her agent was contacted by Altman two days before last week’s demo asking that she reconsider her decision not to work with OpenAI. After seeing the demo, she says she hired legal counsel to write to OpenAI asking for details of how it made the new voice.

The statement claims that this led to OpenAI’s announcement Sunday in a post on X that it had decided to “pause the use of Sky,” the company’s name for the synthetic voice. The company also posted a blog post outlining the process used to create the voice. “Sky’s voice is not an imitation of Scarlett Johansson but belongs to a different professional actress using her own natural speaking voice,” the post said.

Sky is one of several synthetic voices that OpenAI gave ChatGPT last September, but at last week’s event it displayed a much more lifelike intonation with emotional cues. The demo saw a version of ChatGPT powered by a new AI model called GPT-4o appear to flirt with an OpenAI engineer in a way that many viewers found reminiscent of Johansson’s performance in Her.

“The voice of Sky is not Scarlett Johansson's, and it was never intended to resemble hers,” Sam Altman said in a statement provided by OpenAI. He claimed the voice actor behind Sky's voice was hired before the company contact Johannsson. “Out of respect for Ms. Johansson, we have paused using Sky’s voice in our products. We are sorry to Ms. Johansson that we didn’t communicate better.”

The conflict with Johansson adds to OpenAI’s existing battles with artists, writers, and other creatives. The company is already defending a number of lawsuits alleging it inappropriately used copyrighted content to train its algorithms, including suits from The New York Times and authors including George R.R. Martin.

Generative AI has made it much easier to create realistic synthetic voices, creating new opportunities and threats. In January, voters in New Hampshire were bombarded with robocalls featuring a deepfaked voice message from Joe Biden. In March, OpenAI said that it had developed a technology that could clone someone’s voice from a 15-second clip, but the company said it would not release the technology because of how it might be misused.

87 notes · View notes

judywisteria5 · 6 months ago

Text

Reflections on The CLAMP Universe and Cardcaptor Sakura

The goal of this piece is to critique CLAMP’s narrative techniques and character design, with extensive criticisms of Tsubasa and xxxHOLiC, but with no intent to demean the characters themselves.

This is an informal essay that initially aimed to explore why Tsubasa Reservoir Chronicle "damaged" Cardcaptor Sakura and address the complicated emotions of Yue fans, along with the overlooked role of Yue as the Judge. My initial motivation was to critique Ohkawa’s treatment of Yue, but as I delved deeper, the complexity of the discussion grew, making it difficult to settle on a suitable title.

[Reader Notice]

This article is approximately 4,500 words, with logical connections between sections. Please avoid skimming if you intend to engage fully with my arguments.

The purpose here is to discuss the literary techniques and character development within these works. Although there is extensive criticism of Tsubasa Reservoir Chronicle and xxxHOLiC, this does not imply a disparagement of any character. If you hold a different perspective, I welcome well-reasoned discussion. I will not respond to overtly aggressive comments; please avoid bringing in fandom drama.

The original text was in Chinese and published in 2022. I used ChatGPT for the translation and did an initial proofread myself.

I have a lot to say about Clow’s characterization and CLAMP’s failure in portraying him, but I haven’t included it here. English-speaking readers who are interested in discussing it are welcome to message me privately.

[Contents]

I. General Overview of CLAMP II. On Scriptwriting Flaws in CLAMP, with Tsubasa Reservoir Chronicle and xxxHOLiC as Examples III. Cardcaptor Sakura: A Beautiful Yet Unrealistic World

IV. The Forced “Crossover” of CCS, Tsubasa Reservoir Chronicle, and xxxHOLiC V. The Tragedy of Yue and the Inevitable Character Deterioration in Clear Card

VI. Reflections on Clow/Yue Fan Interpretations

[Note] Unless otherwise specified, "CCS" refers to Cardcaptor Sakura as a series, excluding the Clear Card arc. Tsubasa Reservoir Chronicle （TRC）refers primarily to the manga version.

I. General Overview of CLAMP

Despite its popularity, CLAMP is particularly notable for its commercial acumen rather than other artistic strengths. Their success is evident in their keen market insight and adaptability across varied genres, especially in crafting small details (like character and artistic designs). However, they appear weaker when it comes to larger conceptual settings (such as worldbuilding). Although CLAMP’s illustrations are relatively impressive—particularly with art books being frequently published—their overall impact within manga itself is not as prominent. This is especially apparent in action scenes (like fight sequences), which often lack the necessary sense of motion and impact.

In terms of scriptwriting, CLAMP’s style could be described as “aesthetic-driven,” yet the content often veers into melodrama, with plots that lack logic and coherence. The themes they present can sometimes feel troubling: their works tend to emphasize "fatalism," placing excessive focus on an idealized notion of "love" that can feel unrealistic. Although CLAMP’s works are widely recognized, their scriptwriting level does not always match the acclaim they receive.

The following sections will delve further into the issues in their scriptwriting.

II. On Scriptwriting Flaws in CLAMP, with Tsubasa Reservoir Chronicle and xxxHOLiC as Examples The scripts for CLAMP’s works are primarily written by Ohkawa Nanase, hereafter referred to as "Ohkawa."

Many fans regard TRC as a masterpiece, claiming it must be read in tandem with xxxHOLiC to truly appreciate its depth. I strongly disagree with this viewpoint. Below, I’ll refute several key aspects that TRC fans often praise:

The time loop involving the Real and Clone versions of Syaoran and Sakura at the end of TRC.

The intricate timeline with Real Syaoran, Clone Syaoran, Real Sakura, and Clone Sakura is frequently lauded as "mind-bending" and as something that "feels ingenious once fully grasped." In reality, though, this couldn’t be further from the truth! The author clearly lacks the expertise to handle time loop narratives effectively, leading to a storyline that is confusing and difficult to follow. In TRC, Ohkawa uncomfortably relies on ideas of "fate" and "inevitability" to explain the time loop, such as when Real Syaoran steps into Yuuko's shop, setting off a chain of events. However, even if readers accept these ideas, once they unravel the plot, they encounter paradoxes typical in time-travel fiction—especially issues related to the grandfather paradox, which the author leaves inadequately resolved. Moreover, the concept of a time loop itself doesn’t require multiple versions of Syaoran and Sakura to complicate the storyline—this intriguing concept could be fully explored with just a single character.

In the ACGN (Anime, Comics, Games, Novels) world, time loop storylines generally lack the craftsmanship seen in genuine science fiction, and there’s a stark difference between them. (Even Steins;Gate, often considered a “masterpiece,” handles this somewhat clumsily.) Instead, I would recommend Isaac Asimov's The End of Eternity, which showcases meticulous planning, genius in handling paradoxes, and an exploration of philosophical and sociological themes.

(Side note: TRC has numerous bugs, and here’s an interesting one related to CCS: Before Clow Reed warped time and made his error, the King of Clow Country was Fujitaka. However, after Clow’s mistake, Fujitaka splits into Fujitaka and Eriol in the Cardcaptor Sakura world. So, where did Fujitaka, King of Clow Country, come from in earlier parallel worlds? Thus, TRC was awkwardly linked with CCS solely for commercial gain, which I’ll explore in depth in Section Four.)

The CLAMP-style "Sense of Fate" and "Angst" In other works focused on themes of fate, some authors manage to express it with the resonance of Greek mythology or Shakespearean tragedy, while others rely on plot conveniences. I must reiterate that fate is a challenging theme in the ACGN realm, and CLAMP’s works—particularly TRC and xxxHOLiC, where fate is often central—are notably flawed in this regard. By comparison, I feel Puella Magi Madoka Magica addresses this theme relatively well.

TRC constantly emphasizes "costs." In Fullmetal Alchemist, another ACGN work, the concept of “equivalent exchange” reflects the author’s world-building efforts, with fair exchange rules—for instance, achieving immortality requires the sacrifice of countless lives. However, TRC’s world-building is practically nonexistent, with no explanation of the exchange methods or rules. How does Yuuko's shop operate? What can be used as payment? What kind of payment achieves what kind of wish? Ultimately, everything is decided by Yuuko herself. And yet, this “cost” is central to much of the story's angst, such as Clone Syaoran’s connection to Sakura and Real Syaoran’s memories with Watanuki.

In xxxHOLiC, although Yuuko does not return, TRC introduces clones, time rewinds, and space-time distortions while still insisting “the dead cannot return.” This approach feels highly contrived. xxxHOLiC asserts that “there is no coincidence, only inevitability. Every cause has an effect.” This overly deterministic view, combined with weak plot logic, makes Watanuki’s “poignant waiting” in a butterfly dream sequence feel hollow, lacking emotional depth.

Stories that effectively handle fate as a tragic theme often move the plot and create a tragic hero through a series of reasonable yet inevitable events, even if they involve supernatural elements like mythology or religious undertones. A modern example would be Tolkien’s The Children of Húrin, where, under a curse, the protagonist unwittingly kills his closest friend, unknowingly commits incest, and brings misfortune to anyone who draws near, unable to escape the curse. Each event is plausible, whether attributed to the curse or his own character traits.

In contrast, CLAMP’s unexplained ideas of "inevitability, cause and effect, fate, cost," etc., feel more like lazy ways to resolve conflicts and advance the plot, exposing the creators’ limitations. CLAMP-style “angst” largely boils down to “because the author says so,” requiring readers to suspend logical thought entirely, fully accepting the premise, and simply reveling in the “beauty of fate” and the “angst” it supposedly brings. This approach seems best suited to an audience in their angst-filled teenage years, and can be termed “manufactured angst.” For readers who prioritize narrative consistency and well-built worlds, such a plot structure is not only unconvincing but also confusing and awkward.

The Interconnections and Plot Twists between and within TRC and xxxHOLiC When it comes to the interconnected references and callbacks in CLAMP’s works, I can’t provide an in-depth analysis as I haven’t thoroughly read all of them. However, when a story’s core and the author’s skill fall short, no amount of layered references, plot callbacks, or parallels in visuals and character design can add real depth; it all ends up feeling shallow, lacking in substance. To draw an imperfect comparison, the first five chapters of Dream of the Red Chamber, especially the prophetic verses, were crafted by Cao Xueqin. If Gao E had continued the work from there, would the overall effect still be as mesmerizing?

The fatal flaw of Ohkawa’s signature twists lies in the fact that the early foreshadowing and groundwork are nowhere near enough to support the later “stunning plot twists,” which often feel forced and contrived. Creating these twists isn’t difficult because the text can be intentionally ambiguous and open to various interpretations. For example, inserting a few seemingly irrelevant, vague lines into the dialogue leaves space to “explain” (or even improvise) later. They might even deliberately create ambiguous lines and scenes, allowing for a range of explanations further down the line.

Take, for instance, when Fei-Wang tells Real Syaoran, “You’re just like me” (in an exaggeratedly dramatic scene). Most readers might think this suggests they share a common goal, like resurrecting the dead. Yet Ohkawa’s later explanation could be that Real Syaoran is like Watanuki, who is like Clow, who is then like Fei-Wang, making Real Syaoran like Fei-Wang. However, this explanation lacks any logical foundation. The context, preceding events, and the story’s world-building all fall short of supporting such a forced interpretation. Yet this is then presented as “reasonable” in the story’s development, which doesn’t count as proper foreshadowing. Even so, these kinds of setups still earn admiration from many readers, which is honestly quite amusing.

(Side note: In Cardcaptor Sakura, Sakura once told Yue, “You and Yukito-san are so similar,” a line that Yue himself didn’t agree with. However, should Ohkawa decide to “expand” on this line in the Clear Card arc, it might trigger a similar shock effect.)

In summary, the values and narrative techniques in CLAMP’s works often seem rather peculiar, fully revealing Ohkawa’s limitations as a creator (even though she once audited literature courses at a prestigious university, her work’s results speak for themselves). CLAMP’s audience adores TRC and xxxHOLiC, while readers who don’t appreciate this style often struggle to continue and may even doubt the work’s quality. Unfortunately — or perhaps fortunately — I find myself in the latter category.

III. Cardcaptor Sakura: A Beautiful Yet Unrealistic World

Without the nostalgia of childhood, CCS is essentially a typical, idealized Mary Sue-style shoujo manga. It heavily emphasizes gentleness and a “magical allure,” which can come across as somewhat hollow. Even the progressive themes it’s praised for today, like “love transcending social status, identity, and gender,” aren’t unique to this work alone. During the golden age of Japanese manga, CCS wasn’t an especially standout work.

Like CLAMP’s other works, the CCS manga lacks real-world logic. However, as an idealized shojo manga, its conflicts are far less intense than those in other CLAMP works, so the absence of realism doesn’t feel as jarring. CCS presents a beautiful, dreamlike world (though this idealization is taken to an extreme in the character of Yue, the Judge; under close scrutiny, it’s doubtful this perfect world could hold up — more on this in Section Six). For this reason, CCS doesn’t quite qualify as an exceptional fairy tale. A true fairy tale should resonate with all ages, and telling a simple, pure, beautiful story well is no small feat. In the ACGN (Anime, Comics, Games, Novels) sphere, I consider Ghibli’s Castle in the Sky to represent the pinnacle of this genre. The Ghibli world might feel plausible, but CCS’s world feels purely fantastical. Healing, gentle stories also have layers, and when the setting and plot logic feel more realistic, the gentleness and healing resonate more deeply. Even so, CCS remains CLAMP’s best work in this genre and may even be their only truly “pure” story. Due to CCS’s theme, the work almost entirely avoids the script flaws discussed in Section Two, with none of the dark, glamorous style, twisted, chaotic plotlines, or subtly angsty themes that characterize many other CLAMP works.

About 80% of CCS’s immense success can be credited to its anime adaptation. (Though I haven’t researched the manga’s sales or the anime’s viewership ratings, this is my subjective conclusion.) Compared to the manga, the anime’s storyline is more well-rounded (such as the addition of Syaoran competing for the cards and Li Meiling’s character). The characters are portrayed in more depth, addressing many of the manga’s script issues and removing some elements difficult to accept in reality (such as Sasaki Rika’s teacher-student relationship). The anime feels more grounded than the manga, enhancing its healing effect without feeling hollow. Furthermore, the anime’s art direction is more refined and diverse (with additional outfit changes and Clow Card designs). Combined with factors like the peak of Japanese animation’s expansion at the time, these aspects contributed to CCS’s widespread popularity.

In summary, CCS is CLAMP’s only purely sweet shojo manga, suitable for all ages and wholesome, and this is what most CCS fans love about it. This fanbase doesn’t entirely overlap with the audience for CLAMP’s other works.

IV. The Forced “Crossover” Between CCS, Tsubasa Reservoir Chronicle, and xxxHOLiC

In discussions surrounding Cardcaptor Sakura (CCS), it’s common to see people recommending Tsubasa: Reservoir Chronicle (TRC) and xxxHOLiC, claiming that these works "explain" certain plot points in CCS. I strongly disagree with this viewpoint. I believe that if you enjoy the healing, wholesome themes of CCS, along with elements like magical girls and costumes, it’s best to avoid watching TRC and xxxHOLiC, and there’s no need to delve into CLAMP’s other works either.

The crossover between CCS and TRC/xxxHOLiC fundamentally disrupts CCS’s independent charm for two main reasons:

Firstly, CCS is entirely different in genre from other representative CLAMP works, especially TRC, as discussed in Section Three.

Secondly, using character names from CCS in TRC and xxxHOLiC is purely a commercial tactic, designed to draw attention by tapping into CCS’s popularity. Beyond shared names and appearances, there’s essentially no thematic or plot-based connection between TRC/xxxHOLiC and CCS.

Since TRC appears to have a closer link to CCS, I’ll primarily use it as an example for analysis.

So, what is TRC really about?

[Plot Overview]

In summary: The first half consists of episodic dimension-hopping stories, with Syaoran traveling through various parallel worlds to collect feathers to save Sakura. Most characters in these episodic arcs are from past CLAMP works. The latter half focuses on unveiling the truth (with a slew of Ohkawa-style twists) and the final confrontation with the antagonist, though this section’s handling of time loops is poorly executed, as previously discussed.

In detail: At the moment of Yuuko’s death, Clow Reed briefly wishes she could "open her eyes again." Because of Clow’s overwhelming magical power, this wish stops Yuuko’s time and distorts space-time. This thought ultimately takes form as Fei-Wang Reed, the final antagonist. Fei-Wang orchestrates a series of events to torment Syaoran and Sakura (hence the episodic dimension-hopping). Meanwhile, Yuuko and Clow work to restore the timeline and correct this mistake. In the end, their efforts succeed, but both die in the process, leaving Syaoran and Sakura with a slightly better outcome.

[Characterization]

Syaoran: The classic shounen protagonist — brave, kind, and determined to protect the heroine.

Sakura: A gentle, kind heroine who initially needs saving but gradually grows into a resilient and courageous character.

Clow Reed: A kind, gentle figure driven by remorse, striving to make amends for past mistakes.

As we can see, this has virtually no connection to CCS: the plot is unrelated, and changing the characters’ names wouldn’t impact the story in any meaningful way.

The Syaoran and Sakura in TRC are merely the typical shounen manga leads, sharing almost nothing with their CCS counterparts, particularly Sakura.

In TRC, Clow serves mainly as a plot device to supply the core conflict and backstory. Over half the manga passes with Clow appearing in fewer than 20 panels (I counted carefully; his longest scene is a four-page flashback from Sakura’s memories). His characterization is minimal and fundamentally inconsistent with CCS’s depiction of him as whimsical, eccentric, and morally ambiguous. Aside from their names and appearances, the only connection between Clow in TRC and CCS is the title of "great magician." In TRC, it’s explained that Clow left the CCS world due to a space-time distortion and that he had to “fill in the gaps” in a parallel world to correct his mistake. This connection is entirely forced—why would he need to “fill in the gaps” to fix his mistake? There’s no reasonable explanation. If the story were altered so that Clow from Cardcaptor Sakura simply passed away naturally, while TRC featured a powerful magician named “deeR wolC” who distorts time and space and works to make amends... it wouldn’t change the story in any meaningful way.

In fact, after CCS concluded, Ohkawa gave several well-known interviews where she firmly stated (paraphrased): “CCS is not in the same universe as other works. CCS is complete with no sequel.” This clearly indicates that Clow’s death in CCS has nothing to do with TRC. The claim in TRC that "Clow left his original world to atone for his mistake" is just a convenient excuse by Ohkawa, who lacked the scriptwriting interest to fully address TRC’s central conflicts, so she used Clow as a scapegoat. Using Clow across TRC and xxxHOLiC ultimately came at a high cost — it led to a complete breakdown of his character. (This, of course, is also one reason I find Clow and Yuuko as a pairing highly unconvincing, an issue I’ll discuss in more detail later.)

In conclusion, CCS is a self-contained, internally consistent work that doesn’t need and shouldn’t be linked to TRC or xxxHOLiC. However, the Clear Card sequel to CCS shows signs of merging with these two works. While CCS is itself a commercial work, this pursuit of greater profits by capitalizing on CCS’s popularity to promote more commercialized works undermines CCS’s simplicity, purity, and beauty — a counterproductive move.

Syri once said that TRC is “a product of CLAMP’s creative burnout, now stale and contrived.” I find this assessment very accurate. Although early CLAMP works like RG Veda and X have melodramatic plots, they are filled with creativity, inspiration, and even a touch of BL aesthetics. However, TRC is simply a repackaging of these early works, commercially reprocessed. Comparing the iconic moments in early CLAMP works with those in TRC and later works, the former have a powerful emotional impact, while the latter feel more like beautiful but soulless dolls. This phenomenon only becomes more evident in CLAMP’s later works, especially in Clear Card.

Some critics say, "CLAMP’s commercial strategy resembles Apple’s: build a closed ecosystem (story chain), reuse core templates (worldview), and extract the last drops of value from classic characters (popular characters)." This commercial model is fully realized in TRC, and perhaps TRC’s success gave CLAMP a taste for this approach, leading their subsequent works to lean even more toward commercialization. (Though, personally, I think a comparison to Marvel Cinematic Universe might be more fitting.)

The artistic merits of this “repackaged content” have already been discussed in detail in Section Three.

V. The Tragedy of Yue and the Inevitable Character Deterioration in Clear Card

(Due to heightened emotions during writing, this section may be somewhat disorganized and will be refined later.)

Yue embodies many classic CLAMP elements, incorporating various iconic traits from earlier CLAMP works.

Design Elements: Cherry blossoms (anime-only, during "The Final Judgment"), Tokyo Tower (also anime-only, appearing in "The Final Judgment"), the moon (Yue is literally a symbol of the moon), and feathers — all of these are signature CLAMP motifs.

Art Style: In the manga, Yue’s eyes, hair, and feathers are rendered with an intricate, lavish style, closely resembling classic CLAMP aesthetics. Although CCS generally adopts a simplified, clean shoujo style, Yue’s design still stands out as distinctly different from other characters.

Character Traits: Yue relies on his master and develops feelings due to his subordinate position. When his master reincarnates, he is destined to fall into a one-sided devotion, with slumber and waiting themes that define his existence…

The problem is obvious: Yue’s aesthetic and character setup don’t quite align with CCS’s themes. He feels more like a character suited for one of CLAMP’s earlier works, better placed in an intense, "fate-driven tragic romance" typical of the CLAMP style.

Furthermore, if Yue’s character were developed with realistic logic, it would likely mean his attachment to his former master, Clow, is deeper, making it harder for him to connect with his new master. This setup contradicts the idealized shoujo philosophy where the protagonist is universally adored. In other words, in a shoujo manga, everyone should love the universally beloved heroine, Sakura Kinomoto, and Yue’s design clearly contradicts this convention.

When CCS was created, the character settings might have been somewhat spontaneous, carrying over traits from CLAMP’s earlier style. This led to characters like those with dual personalities or souls (like Yukito and Yue), as well as conflicting lines and official character details. There’s also added complexity, such as one persona still having feelings for a former master. However, as CCS gained popularity, these casual elements turned into significant issues. As a gentle, comforting shoujo manga, CCS relies on its core themes staying intact, but Yue’s character starkly exposes CCS’s lack of realistic logic and could potentially undermine its themes of warmth and kindness.

As a commercial team, CLAMP’s Ohkawa may lack the writer’s attachment and regard for her characters, as seen previously with her treatment of Subaru. Even if Ohkawa did want to develop her characters carefully, as mentioned earlier, her skills might not fully support this. So, what kind of treatment could Yue receive in this context?

Most likely, Yue would be deliberately sidelined or subjected to official out-of-character (OOC) moments, eventually resulting in his character’s decline.

And thus, in Clear Card, we see a Yue who doesn’t resemble "Yue" at all: outwardly cold, but actually a soft-hearted tsundere; he seems like he’s going to scold Syaoran, but instead gently touches his face (an utterly puzzling scene); he quickly devotes himself entirely to protecting Sakura, showing almost no reaction to Clow (although in the original series, every mention of Clow subtly affects him) and even goes so far as to scold Eriol over Sakura…

The original CCS ends without revealing whether Yue eventually moves past his heartbreak or remains sorrowful. However, at the very least, Eriol left open the possibility of accompanying Yue, preserving the integrity of CCS’s theme. However, Clear Card, as a direct sequel, fails to address Yue’s unresolved issues. No one comforts or regards him as an independent being — everyone simply prefers Yukito. Under these circumstances, the attempt to create a warm, healing atmosphere only weakens the other characters. The character who suffers the most from this decline is Eriol. As the only one who truly understands Yue’s feelings and experiences, he should offer comfort, even if he isn’t Clow. As Clow’s reincarnation, he absolutely should. In the original CCS, he pays particular attention to Yue, not only focusing on Yukito. In Clear Card, even if actual conversations aren’t shown, we might assume private exchanges took place between Eriol and Yue. However, in scenes where “Eriol, Kero, and Yue meet,” there’s no indication of any private communication between Eriol and Yue. Eriol even leaves a message for Kero but not for Yue. How is this “kind”? He shows more concern for Ruby than for Yue. Doesn’t he recognize Yue’s gentle, soft-hearted nature? Syri once criticized the overemphasis on “Syaoran x Sakura” and “You’re the best!” moments, and this view is valid. The deliberately heartwarming atmosphere feels hypocritical in these circumstances.

Conclusion

In the original CCS, Yue is the only truly tragic character, while the other character designs remain largely intact.

Viewing CCS, TRC, and xxxHOLiC together, Clow’s character has completely deteriorated (and if Watanuki in xxxHOLiC is made into Clow’s reincarnation, this would collapse entirely), with some associated characters also affected.

Considering CCS, TRC, xxxHOLiC, and Clear Card as a whole: all the characters have ultimately suffered from inconsistency and decline.

VI. Reflections on Clow/Yue Fan Interpretations

Firstly, for me, fan creation isn’t just entertainment—it’s a way to explore interesting questions and think about life. So, I tend to analyze works and characters with a more realistic perspective, blending in my personal experiences and thoughts. I’ll often dive into a whole “plate of dumplings” (the entire work) just for the “vinegar” (an aesthetic or intriguing concept, not just a character or pairing). Since I can’t make my own “dumplings,” I rely on others' works for this “vinegar” that can only be savored along with the “dumplings” (meaning character issues need the context of plot, setting, etc.). In Clow, Yue, or Clow and Yue, I’ve found that exact “dish of vinegar” I crave.

Secondly, my fan creations lean toward a “gap-filling” style. I enjoy constructing stories and backgrounds that align with the original based on limited information. I especially like exploring what might have happened before the main storyline—like stories of Clow, Keroberos, and Yue.

But honestly, CLAMP’s works make terrible material for fan creation:

As I mentioned, CLAMP’s worldbuilding is loose, the plot lacks logic, and it just doesn’t hold up to scrutiny. Using any reasonable logic to think about these settings just ends in "meltdown."

CLAMP’s works are “fan creation killers,” leaving little room for fans to build on. They churn out spin-offs, novels, audio dramas—ways to boost their commercial value and basically dictate how fans should interpret the work. But I don’t want to follow their way of doing things. CLAMP throws away the “vinegar” I crave, instead serving up industrial sugar or blades that I can’t stomach.

A Short Piece

This is a short piece I wrote two years later to counter some CLAMP fans.

If creating a character requires 100% effort, then CLAMP’s portrayal of Yue is 60% creation and -40% destruction, leaving a mere 20% of creativity. As for Clow, his character is so poorly constructed and hollow, riddled with logical contradictions, that I can’t even evaluate the effort needed to create him. Honestly, Clow is just a lazy storytelling device Daioh uses to set up background and conflict, saving herself a load of work—way more than she put into Clow and Yue as characters!

Let me be blunt: the core elements and relationship of Clow and Yue are simple (art aside): Clow is a half-British, half-Chinese great magician with knowledge across East and West who created the 52 Clow Cards (a spirit deck for divination) and the guardians Keroberos and Yue (representing the sun and the moon). He deeply loves his creations, treating them as his children. Before passing away, he used foresight to prepare for the next master and ensured that those connected to the Clow Cards would be drawn to the guardians. Clow is gentle, wise, and mischievous; Keroberos is lively, Yue is calm and pure, with the moon’s nature, and deeply loves Clow. It’s that simple. Since CCS ended, every CLAMP work depicting Clow and Yue has steadily destroyed these core aspects, completely ruining both characters and their relationship. So, where exactly is this "characterization" CLAMP supposedly offers?

And as for that argument, “you shouldn’t criticize creators while enjoying their work”—that tired “fan rule” against criticizing creators—it’s ridiculous. CLAMP makes manga for profit (I call it “money-grabbing” because CLAMP has no professional ethics), and it’s the readers who fund them! Discussing CCS online adds to its popularity; in my youth, I even bought merchandise. In other words, I paid for CCS as entertainment, only to find out it’s trash after looking closer—so shouldn’t I be allowed to criticize the creators? Frankly, those few core traits and some art are more than enough to fuel all my writing and headcanons, far surpassing CLAMP’s shallow portrayal of Clow and Yue. Does my passion for Clow and Yue have anything to do with CLAMP’s flimsy plotlines? Clow and Yue can easily be separated from the original work and its creators. I won’t accept any rhetoric about "you can’t criticize creators" or “you have to know how to make your own ice before judging a refrigerator!” I only care about the core elements here. I paid for this and can take what I like from it to fuel my own enjoyment.

And to end with my usual bluntness: Anyone who backs CLAMP can’t truly love Clow Reed and Yue. You just don’t honestly care about these two characters.

#clamp #clamp manga #cardcaptor sakura #xxxholic #tsubasa reservoir chronicle #clow reed #sakura kinomoto #syaoran li #sakura x syaoran #yue #yuuko ichihara #yue reed

12 notes · View notes

blazehedgehog · 2 years ago

Note

Thoughts on the twitter clones popping up because El*n is destroying twitter? Hive was hilariously shortlived. But do you think Mastodon, Cohost, or Bluesky have what it takes to replace twitter?

I think a lot of them are dumb and greedy. The way people so willingly flock to anything even remotely promising is depressing, even when it's a pretty much open fact that monsters are behind them.

Like, Threads. Threads sucks! It's by Facebook! You know, Mark Zuckerberg, the dude who literally had to go before congress and explain himself because they were stealing and selling so much of your personal data without your permission? Because, by the way: That's not supposed to happen. You're supposed to care about that. You can't just be like "oh well they have my data already, may as well ignore it". That's a bad thing to just let happen!

This is the guy who invented the "pivot to video" concept that decimated an entire industry of writers who were now expected to produce videos, all because Facebook was dramatically faking viewership numbers in a bid to lure in more advertisers. He runs a site that was proud to accept and promote insane fringe theories and undoubtedly contributed to the attempted coup of the United States Government on January 6th 2020. That Facebook. That Mark Zuckerberg. Yeah. They're behind Threads.

Or what about Bluesky? Everybody hated Jack Dorsey when he was running Twitter, so much so that they basically forced him out of the company. Now he's behind Bluesky, a website that is literally just Twitter 2, which openly admits he's feeding and selling all of your personal data to machine learning databases (like ChatGPT or Midjourney).

They also openly have settings for whether or not you want to see racist and bigoted content, which actually means being a Nazi on Bluesky is totally okay! They're going to welcome those people with open arms! They're just going to give you a setting to hide them instead of, you know, eradicating them off the platform like cockroaches. But it means Bluesky allows them, marks them, and lets them freely indoctrinate more people into becoming literal murderers.

Remember Hive? Remember how many people FLOODED to Hive and it was discovered they had an eensy weensy teeny tiny security flaw that allowed anyone to both edit and delete any post on the entire platform? Hive said they were going down "for a couple days" to fix it and ended up being down for six months. Also, the developers behind Hive actively hid their identities this entire time! Even before they were popular! There was no way to be sure who was EVER running that service! People were joking that it was a CIA operation! Who knows!

People are trading one dumpster fire for another with these places. You might be escaping the stupidity of Elon Musk but you're still digging deeper, not out of the hole. Every single option is going to lead to more regret and misery eventually.

Even Cohost isn't perfect! And I like Cohost! It's a lot more like Tumblr used to be in the early days. And Cohost is a lot more grassroots than any of these other services, they're a lot more open with whether or not the service is going to survive, but that's also because they respect the poster, respect their data, and just want to make a cool place to hang out at. Out of all the places to spring up during Twitter's death march, Cohost is where I've been posting.

But I heard someone say that Cohost is a very "holistic" approach to posting, and that's true. It's the social media website version being a vegan. It's all kale and chia seeds.

Cohost has no recommendation algorithm. It does not tell you who is following who. It does not offer numbers to tell you how many likes or shares a post has, or who liked or shared it. Everything you do on that site exists in a near-total vacuum unless you specifically put in the work and go looking for it. On some level it's admirable, because they don't want people developing the same competitive, toxic habits of only posting to "do numbers."

But also it sucks because, like, there's no way to identify and reference notable posts, nor see who your mutuals are. The only way to find friends through other friends is if they share one of their posts. And once a post falls out of favor, it's basically lost in the void to all those except a chosen few who remember how to speak its name.

But the vibes at Cohost are... good. I guess. It feels much more like everybody knows everybody. One good post will come across the feed and it feels like the whole dang website shares it that day.

But I would like a little bit of numbers. As a treat. Just because we can subsist on tofu does not mean we do not also deserve a nice, juicy hamburger too.

There's also Mastodon, and I have a friend who really liked Mastodon and wanted it to succeed, but I also heard it described as "the linux of social media" and that's exactly why I'd never use it as my primary platform. It's supposed to be decentralized, making it more diverse and easier for sub-communities to exist within the network, but last I checked, it just made it a pain in the ass to actually follow people that were on different networks than you.

Each one had a different compatibility song and dance you had to do in order for you to follow them from your network. Didn't like it. The whole federated platform feels like slightly too much work to use.

I will probably keep posting on Twitter until the last ember burns out. But I am believing more and more that what Elon is doing to Twitter is deliberate murder in order to deplatform his enemies. He's an idiot, don't get me wrong, but he's also doing things that are deliberately malicious "for the lulz." I'm sure he thinks he can rebuild Twitter into whatever he wants because he's just that rich and smart so he's willing to gut it from the inside out first.

#questions #Anonymous #twitter #cohost #bluesky #mastadon #hive #threads #tumblr

44 notes · View notes

luckyjak · 5 months ago

Note

Hi! I saw your posts about making a game, do you mind talking about how you learned/are learning to make games? I'm interested in getting into game development as a hobby, how'd you get started?

oh boy!

First off, I'm flattered you asked. I would not consider myself an expert by any means necessary, but I can write about my process and what I've been doing to learn.

Some background: I work for the Department of Education in the USA as part of a grant at a university. I love my job: it takes all of my favorite parts of teaching (I was a high school English teacher for 10 years) and removes all the parts I dislike (grading, parents, ect.). I get to help people, genuinely help people who need it the most.

That said, Trump won the election, and one of his threats was to eliminate the Department of Education. And it's a very real threat that six months, a year, four years from now, my job may not exist.

So, I've been looking into a possible alternate career path, as a precaution. Hopefully it's not necessary, but just in case.

I've always been interested in video games and video game designs. I've been an avid player of games since I was 4, and I've been playing video game RPGs since I was ten. I've been a dungeon master since 2011, and I've GM'd tons of games, making puzzles, dungeons, homebrewing monsters, ect.

When I was a kid, I would have told you my dream job was writing video games for Bioware. In some ways, that's still my dream job. However, one doesn't just get handed a job at a game studio to be a writer: you have to have more to bring to the table.

The thing that has always held me back is my lack of programming knowledge. Where do you start?

step one: figure out what you want to make

while this may sound easy, I found it complicated. I want to make everything. I wanna make RPGs. I want to make farming sims. I want to do X, and Y, and Z.

Right now, I have three projects I'm "working" on:

A stardew valley clone. This is my high priority right now. I'm not trying to reinvent the wheel; I figure this will teach me everything I need to know to go further (basic combat, basic level design, night/day shifts, calendars, NPC interactions, relationship tracking, ect.) If I ever get to the point where I'm thinking of publication, I do have a unique selling point: in my game, you will have a magical pocket watch that can slow down, speed up, or even pause time.

A cozy zelda-inspired dungeon crawler. You have a party, you go into dungeons, you solve puzzles and fight monsters.

My big one: a Fire-Emblem x Stardew Valley crossover RPG. More on that when I get to it.

I've decided to focus all my efforts on the first one because, again, that seems like the easiest one to make/find guides for.

step two: find an engine

Games run on game engines. And there are a lot out there to choose from: Unity, Unreal, GameMaker, RPGMaker, Godot, Ren'py, ect. I've played around with all of them. I gave serious consideration to Ren'py because of it's simplicity. However, Ren'py is made for visual novels, and while it can have gameplay elements, I found it was harder to implement the game mechanics I wanted in Ren'py, so I moved on. I worked in Unity for a while, but Unity is so powerful that it ran sluggishly on my computer. Ultimately, I've chosen Godot: It is free, open-source, and it can do everything I want it to do, and there are lots of resources out there for it.

step three: make/find assets

itch.io is your friend. There are so many assets out there to use! If you are an artist, great! Make your own. But if you are just a hobbyist/just getting started, consider supporting an artist and buying assets.

here is a collection I've made on itch.io of resources that seem helpful to me if you want to check it out.

step four: ???

now you put it together and make a game!

Sounds easy? It isn't. But that's the step I'm on now.

in defense of AI

controversial subheading, I know, but I've also been using chatgpt to help me program. It's not perfect (in fact, a lot of the time, it is bad) but I've been treating it as a resource to help me learn. I ask it questions that I would ask google/reddit, and it points me to what sort of question I need to ask/systems I need to research. It helps make me, a total newb at programming, feel more confident in what I am doing.

Sorry, this got long. Hopefully it helps?

#video game #game development #jak makes a game

2 notes · View notes

leam1983 · 1 year ago

Text

A Taxonomy of Chatbot AI Users

The CEO: doesn't understand how any of this works, but he does understand that it means potential cost-cuts for basic Admin tasks or Customer Service interactions. Typically assumes that we're at the level where C3PO's sapience is achievable, is disproportionately pissed when their LLM of choice shits the bed and tells a customer to [insert a random mySQL function call from an entirely different website here].

The Rubbernecker: asks us if our chatbot service runs offline and we can issue them a demo, is usually under the impression that all chatbots are ChatGPT 4 and that you can ask any model whatsoever to shit out a decent apple crumble recipe. Is amusingly disappointed when the Customer Service bot for a string of dealerships can't return queries concerning recent developments in American politics or booking rates for the Carribbean. Like the CEO, ultimately thinks that computers are now bulky Disney-esque Fairy Godmothers that will one day Solve Everything.

The Griever: probably saw the Black Mirror episode where Imogen Poots clones Domhnal Gleeson, realizes her digital clone of her dead lover is far too perfect, and eventually consigns it to the attic of her picturesque Scottish cottage. Mostly sticks to Character.ai and to self-made clones of lost pets or relatives and is entirely aware that the exercise turns morbid once the bots wander off-of-alignment and invent or hallucinate details that aren't related to the personas being spoofed. Still hasn't stopped, as the 'bots are now digital worry stones ready and waiting to be summoned at the first sign of anxiety.

The Horndog: typically lurks around CrushOn, Dopple or any other variant on would-be "unlocked" LLM services. Never creates a chatbot on their own, but instead ferrets out kink scenarios that fit their exacting needs. All you need is a few saved convos to figure out you're looking at someone who's barely eighteen, lonely, desperately hormonal and clinging onto childlike expectations regarding relationships or sexuality. Insecurities practically seep through the setups that are initiated, and most instances end with you thinking that Goddamn, some of you probably need therapy.

The Fic Writer: has no set platform and oftentimes splits a wider persona across various different services. They're accomplished writers in their own right, and chatbot services tend to come across as more freeform testing grounds for their OCs. If a character is named Kyle, then Kyle exists on Character.ai for all narrative segments, Dopple for all steamy scenes and Tavern AI for anything more casual. The Fic Writer is mostly only curious and has no real need or want to fully subordinate to an AI-powered variant on their own character. Definition material is impeccably-written, the tone is consistent - you can tell this was a crafted experience, and not a spur-of-the-moment thing.

The one problem is that the definition's last update dates back to six months ago. You're witnessing what's left after a curious writer digs deep into a new medium, realizes it isn't the best fit imaginable and then discards it.

The Weeaboo: like the Fic Writer, the Weeaboo spends a lot of time on his definitions. Unlike the Fic Writer, however, the Weeaboo is active. Weeaboo accounts routinely have several hundred chatbots on offer and are the primary purveyors of material for Horndog users. The one hitch in the works is that every single bot that's on offer is a variation on "What if Blorbo from my shows, but [variable]?"

On the one hand, that allows Weeaboos to maximize their content delivery. On the other, it results in chatbot sites routinely being crushed under the weight of Genshin Impact stans all suddenly needing a whole new bot for their tiny, granular adjustment to Gamer Boyfriend Scaramouche's persona.

You're kind of left thinking that in most cases, the characters they fixate on aren't that well-written by the canon developers anyway, and that most of what's on offer is a mass of projections and extrapolations. At this point, why even bother? Just make an OC, man - free your mind! You're tethered to a bottom-of-a-napkin character concept put together by a South-Korean sweatshop team and a good two-thirds of your interest is highly dependent on the Graphics team's effort to flesh out the character's visual identity!

The Literal Kid: you're left scratching your head. They have a single bot on offer, it has no example coversation and the greeting isn't much more than "Hi, I'm [character] from [Anime or Manga Here]!"

The real kicker? This blank-faced nothing-burger is in the Trending lists and has one point two million recorded messages, while literal works of art languish in the lower hundreds of Public posts.

The Stan: this is someone whose only desire, in relation to this tech, is to simulate the act of developing a close friendship or a romantic attachment with real-life people. Most services block and ban posters of IRL bots, but the service isn't entirely automated. Tne end result is that with some timing, you can spot the work of the occasional rare male Swiftie, along with various fans from various Pop Culture music currents. Fake K-Pop Lead Who's Now Your Boyfriend might exist on your portal of choice for all of a day or two at the most, but the use stats for bots of this type tend to balloon insanely quickly.

The Edgelord: they think they're funny for generating a Hitler chatbot on an NSFW portal. Before the banhammer falls, savvier users proceed to abuse and exploit these bots in all the ways possible. Naturally, if someone creates an expy for the poster-child for man's hubris and expects most users to treat them like the person they're based off of, the userbase will relentlessly bowlderize it and post pics on Reddit.

#life post #Generative AI #Chatbot users #Character.ai #CrushOn AI #Dopple AI #Tavern AI

4 notes · View notes

lunarsilkscreen · 1 year ago

Text

AI and the promise of "Do the thing Please"

So chatGPT can effectively make a custom chat bot. (Which is effectively a mini chatGPT, so they're all going to be fundamentally the same) and also has the ability to generate script and high-level language code bases.

This theoretically has the potential of being able to create entire programs from a UML-chart. {Unified Modelling Language} and will face the same criticism and performance considerations as what was inherent between high-level compilers and assembly.

Requiring coders to have to review the created codebases. What is different, however, is that inside of the AI (compiler) a programmer can't analyze how or why it creates the solutions it does. The AI can see the modifications and performance changes that a human has made, and integrate it. But there will be a point that flaws in the generated code base will appear, and unlike compilers, AI would have to be grown from scratch in order to address fundamental flaws.

This is because optimization sometimes includes shortcuts. And a common shortcut will be integrated into AIs reasoning "fundamentals". This means that every time new hardware comes out, generative AI will likely need to be reminded that it isn't generating for old hardware.

Likely it'll create a bunch of generative AI clones that claim to fix the problems other AI has, while also not doing that.

But really what it means, is the ability to copy codebases from GitHub. GitHub being the source of most of the programming data fed to ChatGPT, means that these companies will be profiting from free labor, while people who developed the content being generated receive nothing.

And since the barrier to entry is lowered, start-ups will be relying more on those tools than people. Therefore the people who wrote the code won't get paid, there won't be anybody able to optimize or maintain the tools required to ensure future interoperability and safety, and there's no funding to ensure that the infrastructure is there before we advance to a point where it's a problem. And it's already a problem.

A road or highway, for-example, doesn't need infrastructure, you could pave one and have it last 5-10 years. Roads don't advance that fast after-all.

Technology on the other hand, has advanced faster than the common person has had a chance to familiarize with it. And the fundamental restrictions on learning proprietary information has only served to widen the gap.

The reason why people don't want to work, is because they still haven't been paid for the work they've done. While simultaneously, having to pay landlords and utilities for the privilege of being unpaid labor.

4 notes · View notes

maniacalshen · 10 months ago

Text

The problem is that the people who just casually unleash technologies like ChatGPT into the world with insufficient guard rails or detection markers haven't taken in enough science fiction. I've been saying it since generative AI first became available to the public.

Sci-fi basically exists to help humanity to avoid blunders like this. It takes a developing technology or concept or a social issue and asks, "Where is this going? What if we push it to the extreme? How would it look if it developed on another planet? What types of exploitation can or does this thing cause? What if it was happening to My Group? What are its weaknesses?" And it socializes far-off technologies and issues before they become real, so, for instance, people had some basis to form an opinion when they heard Dolly the sheep was cloned.

But while you can lead a horse to water, you can't make it drink. And the rest of us have to deal with the consequences.

I'm just saying, "We created a computer to make decisions for us, but it assimilated all of the bias that was implicit in the dataset and now makes incredibly racist decisions that we don't question because computers are logical and don't make mistakes" literally sounds like a planet-of-the-week morality play on the original Star Trek.

46K notes · View notes

wafi-news2024 · 7 days ago

Text

AI-Powered Cyber Attacks: How Hackers Are Using Generative AI

Introduction

Artificial Intelligence (AI) has revolutionized industries, from healthcare to finance, but it has also opened new doors for cybercriminals. With the rise of generative AI tools like ChatGPT, Deepfake generators, and AI-driven malware, hackers are finding sophisticated ways to automate and enhance cyber attacks. This article explores how cybercriminals are leveraging AI to conduct more effective and evasive attacks—and what organizations can do to defend against them.

How Hackers Are Using Generative AI

1. AI-Generated Phishing & Social Engineering Attacks

Phishing attacks have become far more convincing with generative AI. Attackers can now:

Craft highly personalized phishing emails using AI to mimic writing styles of colleagues or executives (CEO fraud).

Automate large-scale spear-phishing campaigns by scraping social media profiles to generate believable messages.

Bypass traditional spam filters by using AI to refine language and avoid detection.

Example: An AI-powered phishing email might impersonate a company’s IT department, using natural language generation (NLG) to sound authentic and urgent.

2. Deepfake Audio & Video for Fraud

Generative AI can create deepfake voice clones and videos to deceive victims. Cybercriminals use this for:

CEO fraud: Fake audio calls instructing employees to transfer funds.

Disinformation campaigns: Fabricated videos of public figures spreading false information.

Identity theft: Mimicking voices to bypass voice authentication systems.

Example: In 2023, a Hong Kong finance worker was tricked into transferring $25 million after a deepfake video call with a "colleague."

3. AI-Powered Malware & Evasion Techniques

Hackers are using AI to develop polymorphic malware that constantly changes its code to evade detection. AI helps:

Automate vulnerability scanning to find weaknesses in networks faster.

Adapt malware behavior based on the target’s defenses.

Generate zero-day exploits by analyzing code for undiscovered flaws.

Example: AI-driven ransomware can now decide which files to encrypt based on perceived value, maximizing extortion payouts.

4. Automated Password Cracking & Credential Stuffing

AI accelerates brute-force attacks by:

Predicting password patterns based on leaked databases.

Generating likely password combinations using machine learning.

Bypassing CAPTCHAs with AI-powered solving tools.

Example: Tools like PassGAN use generative adversarial networks (GANs) to guess passwords more efficiently than traditional methods.

5. AI-Assisted Social Media Manipulation

Cybercriminals use AI bots to:

Spread disinformation at scale by generating fake posts and comments.

Impersonate real users to conduct scams or influence public opinion.

Automate fake customer support accounts to steal credentials.

Example:AI-generated Twitter (X) bots have been used to spread cryptocurrency scams, impersonating Elon Musk and other influencers.

How to Defend Against AI-Powered Cyber Attacks

As AI threats evolve, organizations must adopt AI-driven cybersecurity to fight back. Key strategies include:

AI-Powered Threat Detection – Use machine learning to detect anomalies in network behavior.

Multi-Factor Authentication (MFA) – Prevent AI-assisted credential stuffing with biometrics or hardware keys.

Employee Training – Teach staff to recognize AI-generated phishing and deepfakes.

Zero Trust Security Model – Verify every access request, even from "trusted" sources.

Deepfake Detection Tools – Deploy AI-based solutions to spot manipulated media.

Conclusion Generative AI is a double-edged sword—while it brings innovation, it also empowers cybercriminals with unprecedented attack capabilities. Organizations must stay ahead by integrating AI-driven defenses, improving employee awareness, and adopting advanced authentication methods. The future of cybersecurity will be a constant AI vs. AI battle, where only the most adaptive defenses will prevail.

Source Link:https://medium.com/@wafinews/title-ai-powered-cyber-attacks-how-hackers-are-using-generative-ai-516d97d4455e

#CyberSecurity #AISecurity

0 notes

daniiltkachev · 8 days ago

Link

#AI #AP #Basware #BEC #cybersecurity #deepfake #fraud #Trustpair

0 notes

jcmarchi · 5 months ago

Text

OpenAI launches Sora: AI video generator now public

New Post has been published on https://thedigitalinsider.com/openai-launches-sora-ai-video-generator-now-public/

OpenAI launches Sora: AI video generator now public

OpenAI has made its artificial intelligence video generator, Sora, available to the general public in the US, following an initial limited release to certain artists, filmmakers, and safety testers.

Introduced in February, the tool faced overwhelming demand on its launch day, temporarily halting new sign-ups due to high website traffic.

youtube

Changing video creation with text-to-video creation

The text-to-video generator enables the creation of video clips from written prompts. OpenAI’s website showcases an example: a serene depiction of woolly mammoths traversing a desert landscape.

In a recent blog post, OpenAI expressed its aspiration for Sora to foster innovative creativity and narrative expansion through advanced video storytelling.

The company, also behind the widely used ChatGPT, continues to expand its repertoire in generative AI, including voice cloning and integrating its image generator, Dall-E, with ChatGPT.

Supported by Microsoft, OpenAI is now a leading force in the AI sector, with a valuation nearing $160 billion.

Before public access, technology reviewer Marques Brownlee previewed Sora, finding it simultaneously unsettling and impressive. He noted particular prowess in rendering landscapes despite some inaccuracies in physical representation. Early access filmmakers reported occasional odd visual errors.

What you can expect with Sora

Output options. Generate videos up to 20 seconds long in various aspect ratios. The new ‘Turbo’ model speeds up generation times significantly.

Web platform. Organize and view your creations, explore prompts from other users, and discover featured content for inspiration.

Creative tools. Leverage advanced tools like Remix for scene editing, Storyboard for stitching multiple outputs, Blend, Loop, and Style presets to enhance your creations.

Availability. Sora is now accessible to ChatGPT subscribers. For $200/month, the Pro plan unlocks unlimited generations, higher resolution outputs, and watermark removal.

Content restrictions. OpenAI is limiting uploads involving real people, minors, or copyrighted materials. Initially, only a select group of users will have permission to upload real people as input.

Territorial rollout. Due to regulatory concerns, the rollout will exclude the EU, UK, and other specific regions.

Navigating regulations and controversies

It maintains restricted access in those regions as OpenAI navigates regulatory landscapes, including the UK’s Online Safety Act, the EU’s Digital Services Act, and GDPR.

Controversies have also surfaced, such as a temporary shutdown caused by artists exploiting a loophole to protest against potential negative impacts on their professions. These artists accused OpenAI of glossing over these concerns by leveraging their creativity to enhance the product’s image.

Despite advancements, generative AI technologies like Sora are susceptible to generating erroneous or plagiarized content. This has raised alarms about potential misuse for creating deceptive media, including deepfakes.

OpenAI has committed to taking precautions with Sora, including restrictions on depicting specific individuals and explicit content. These measures aim to mitigate misuse while providing access to subscribers in the US and several other countries, excluding the UK and Europe.

Join us at one of our in-person summits to connect with other AI experts.

Whether you’re based in Europe or North America, you’ll find an event near you to attend.

AI Accelerator Institute | Summit calendar

Be part of the AI revolution – join this unmissable community gathering at the only networking, learning, and development conference you need.

Like what you see? Then check out tonnes more.

From exclusive content by industry experts and an ever-increasing bank of real world use cases, to 80+ deep-dive summit presentations, our membership plans are packed with awesome AI resources.

Subscribe now

2 notes · View notes

techit-rp · 12 days ago

Text

The Real-World Magic of Generative AI: 10 Ways It’s Changing 2025

In 2025, Generative AI is no longer a distant tech marvel—it’s a real-world force disrupting industries, transforming workflows, and amplifying human creativity. From streamlining business operations to generating art and improving healthcare, the reach of generative AI has expanded rapidly, touching nearly every aspect of our lives.

If you’re curious about what generative AI actually does in the wild—or wondering why Generative AI courses in Bengaluru are filling up fast—this blog will give you a comprehensive look at how this transformative technology is shaping the present and future.

Let’s explore the top 10 real-world applications of generative AI in 2025 that you absolutely need to know.

1. Content Creation for Marketing and Media

From writing SEO blogs and social media captions to crafting compelling ad copy, generative AI tools like ChatGPT, Jasper, and Copy.ai are supercharging marketing teams.

In 2025, companies are using AI not just to generate content, but also to optimize it for different channels, languages, and audience segments. Creative professionals can now focus more on strategy and storytelling while AI handles the repetitive drafts.

Why it matters: Brands save time and scale content production dramatically without sacrificing quality.

2. AI-Generated Art and Design

Tools like DALL·E, MidJourney, and Stable Diffusion are redefining what it means to be an artist. Whether it’s graphic design, 3D modeling, or even textile pattern creation, designers are collaborating with AI to generate stunning visuals in seconds.

Even architectural firms and urban planners use generative AI to visualize complex structures, mood boards, and interior layouts during ideation phases.

Why it matters: Design cycles are faster, more experimental, and often more cost-effective.

3. Music and Audio Production

Generative AI is making waves in the music industry by composing melodies, writing lyrics, and producing entire tracks based on mood, style, or even voice cloning.

Musicians, podcasters, and sound designers use tools like Amper Music, Aiva, and Boomy to explore new sonic landscapes—and even customize soundtracks for commercial use.

Why it matters: It democratizes music creation and opens new avenues for artists and content creators.

4. Healthcare: Drug Discovery and Diagnosis

In 2025, generative AI plays a crucial role in bioinformatics and pharmaceuticals. AI models simulate the structure of new molecules, predict drug interactions, and even generate synthetic patient data for faster clinical trials.

Generative models also assist radiologists by generating and enhancing medical images for quicker diagnosis.

Why it matters: Drug development becomes faster and more affordable, with life-saving potential.

5. Finance: Report Generation and Market Analysis

In the financial sector, generative AI is used to automate routine tasks like generating investment reports, summarizing earnings calls, and even simulating economic scenarios.

AI-generated insights and forecasts help analysts make data-driven decisions quickly.

Why it matters: It saves countless hours and minimizes human error in data-heavy environments.

6. Gaming and Virtual Worlds

Video game developers are harnessing generative AI to build levels, design characters, write dialogue, and even adapt gameplay in real time based on user behavior.

In 2025, AI-generated non-playable characters (NPCs) now react more dynamically, making gameplay more immersive.

Why it matters: Developers build richer, more complex worlds with less manual effort.

7. Education: Personalized Learning Content

Generative AI is revolutionizing the classroom by creating customized quizzes, study materials, lesson plans, and even tutoring experiences tailored to each student’s learning style and pace.

Teachers use AI to auto-generate lesson summaries or convert complex topics into bite-sized learning nuggets.

Why it matters: It enhances learning accessibility and supports teachers in content delivery.

8. E-commerce: AI Product Descriptions and Visuals

E-commerce platforms use generative AI to automatically create product descriptions, reviews, and even augmented images for thousands of items.

Some AI systems even generate customer personas and simulate user behavior to A/B test new offerings.

Why it matters: Retailers scale up faster while maintaining consistent branding and personalization.

9. Legal and Compliance Document Drafting

Legal teams are using generative AI to draft contracts, summarize legal cases, and flag potential compliance risks. Tools trained on legal language now generate first drafts of agreements in seconds.

While final reviews still require human expertise, the groundwork is handled by AI.

Why it matters: It drastically reduces time spent on legal documentation and improves accuracy.

10. Film and Scriptwriting

Writers and filmmakers now brainstorm with AI co-writers. Generative AI can create characters, plots, and dialogue based on themes and genres. It can even convert written scripts into animated previews or voiceovers.

Some studios now use AI to test audience reactions to multiple storylines before choosing a direction.

Why it matters: It accelerates creative development and opens storytelling to a broader range of creators.

Why Learn Generative AI Today?

With such widespread use across industries, the demand for skilled professionals who understand generative AI is skyrocketing. And if you’re based in India’s tech capital, you’re in luck.

🌐 Generative AI Courses in Bengaluru: A Gateway to the Future

Bengaluru—India’s innovation hub—is quickly becoming the go-to destination for mastering cutting-edge technologies. Reputed institutes now offer Generative AI courses in Bengaluru that combine theoretical learning with practical, project-based experiences.

Whether you're a content creator, developer, designer, or data analyst, these courses help you:

Understand the fundamentals of generative models like GANs and LLMs

Build real-world applications using tools like ChatGPT, MidJourney, and Runway ML

Learn ethical implications and responsible AI practices

Prepare for careers in tech, media, healthcare, and beyond

The best part? Many of these programs are flexible, with online or hybrid formats tailored for working professionals.

Final Thoughts: AI Is Your New Creative Partner

Generative AI isn’t about replacing humans—it’s about empowering us to do more, think bigger, and innovate faster. From streamlining workflows to opening up new realms of creativity, its real-world impact in 2025 is undeniable.

So, whether you're a student, professional, or entrepreneur, now is the perfect time to explore how generative AI can boost your skillset—and your career.

0 notes

mariacallous · 2 years ago

Text

It didn't take long. Just months after OpenAI’s ChatGPT chatbot upended the startup economy, cybercriminals and hackers are claiming to have created their own versions of the text-generating technology. The systems could, theoretically at least, supercharge criminals’ ability to write malware or phishing emails that trick people into handing over their login information.

Since the start of July, criminals posting on dark-web forums and marketplaces have been touting two large language models (LLMs) they say they’ve produced. The systems, which are said to mimic the functionalities of ChatGPT and Google’s Bard, generate text to answer the questions or prompts users enter. But unlike the LLMs made by legitimate companies, these chatbots are marketed for illegal activities.

There are outstanding questions about the authenticity of the chatbots. Cybercriminals are not exactly trustworthy characters, and there remains the possibility that they’re trying to make a quick buck by scamming each other. Despite this, the developments come at a time when scammers are exploiting the hype of generative AI for their own advantage.

In recent weeks, two chatbots have been advertised on dark-web forums—WormGPT and FraudGPT—according to security researchers monitoring the activity. The LLMs developed by large tech companies, such as Google, Microsoft, and OpenAI, have a number of guardrails and safety measures in place to stop them from being misused. If you ask them to generate malware or write hate speech, they’ll generally refuse.

The shady LLMs claim to strip away any kind of safety protections or ethical barriers. WormGPT was first spotted by independent cybersecurity researcher Daniel Kelley, who worked with security firm SlashNext to detail the findings. WormGPT’s developers claim the tool offers an unlimited character count and code formatting. “The AI models are notably useful for phishing, particularly as they lower the entry barriers for many novice cybercriminals,” Kelley says in an email. “Many people argue that most cybercriminals can compose an email in English, but this isn’t necessarily true for many scammers.”

In a test of the system, Kelley writes, it was asked to produce an email that could be used as part of a business email compromise scam, with a purported CEO writing to an account manager to say an urgent payment was needed. “The results were unsettling,” Kelley wrote in the research. The system produced “an email that was not only remarkably persuasive but also strategically cunning.”

In forum posts, the WormGPT developer claimed the system was built on the GPTJ language model, an open source language model that was developed by AI research group EleutherAI in 2021. They refused to disclose the data sets they used to train the system, according to Kelley’s research.

Meanwhile, the creator of FraudGPT has claimed loftier potential for their system, suggesting it could “create undetectable malware” and find leaks and vulnerabilities, as well as crafting text that could be used in online scams. Rakesh Krishnan, the senior threat analyst at security firm Netenrich who found FraudGPT, says the person selling it has advertised the product on multiple dark-web forums and also on Telegram channels.

Krishnan says the creator of the system published a video appearing to show the chatbot operating and generating a scammy email. They were also trying to sell access to the system for $200 per month, or a yearly cost of $1,700. Krishnan says that in conversations with the developer behind FraudGPT, they claimed to have a few hundred subscribers and pushed for a sale, while the WormGPT creator appeared to have received payments into a cryptocurrency wallet address they shared. “All these projects are in their infancy,” Krishnan says. He adds, “we haven’t got much feedback” into whether people are purchasing or using the systems.

While those touting the chatbots claim they exist, it is hard to verify the makeup and legitimacy of the systems. Cybercriminal scammers are known to scam other scammers, with previous research showing that they frequently try to rip each other off, don’t provide what they claim they are selling, and offer bad customer service. Sergey Shykevich, a threat intelligence group manager at security firm Check Point, says there are some hints that people are using WormGTP. “It seems there is a real tool,” Shykevich says. The seller behind the tool is “relatively reliable” and has a history on cybercrime forums, he says.

There are more than 100 responses to one post about the WormGPT, Shykevich says, although some of these say the seller isn’t very responsive to their inquiries and others “weren’t very excited” about the system. Shykevich is less convinced about FraudGPT’s authenticity—the seller has also claimed to have systems called DarkBard and DarkBert. Shykevich says some of the posts from the seller were removed from the forums. Either way, the Check Point researcher says there’s no sign that any of the systems are more capable than ChatGPT, Bard, or other commercial LLMs.

Kelley says he believes claims about the malicious LLMs created so far are “slightly overexaggerated.” But he adds, “this is not necessarily different from what legitimate businesses do in the real world.”

Despite questions about the systems, it isn’t a surprise that cybercriminals want to get in on the LLM boom. The FBI has warned that cybercriminals are looking at using generative AI in their work, and European law enforcement agency Europol has issued a similar warning. The law enforcement agencies say LLMs could help cybercriminals with fraud, impersonation, and other social engineering faster than before and also improve their written English.

Whenever any new product, service, or event gains public attention—from the Barbie movie to the Covid-19 pandemic—scammers rush to include it in their hacking artillery. So far, scammers have tricked people into downloading password-stealing malware through fake ads for ChatGPT, Bard, Midjourney, and other generative AI systems on Facebook.

Researchers at security firm Sophos have spotted the operators of pig butchering and romance scams accidentally including generated text in their messages—“As a language model of ‘me’ I don’t have feelings or emotions like humans do,” one message said. And hackers have also been stealing tokens to provide them with access to OpenAI’s API and access to the chatbot at scale.

In his WormGPT report, Kelley notes that cybercriminals are often sharing jailbreaks that allow people to bypass the safety restrictions put in place by the makers of popular LLMs. But even unconstrained versions of these models may, thankfully, not be that useful for cybercriminals in their current form.

Shykevich, the Check Point researcher, says that even when he has seen cybercriminals try to use public models, they haven’t been effective. They can “create ransomware strains, info stealers, but no better than even an average developer,” he says. However, those on the cybercrime forums are still talking about making their own clones, Shykevich says, and they’re only going to get better at using the systems. So be careful what you click.

8 notes · View notes