Character.AI Unveils AvatarFX: A Groundbreaking AI Video Generation Model

Character.AI unveils AvatarFX, an AI video model to create lifelike chatbots | TechCrunch

Character.AI, a leading platform for chatting and roleplaying with AI-generated characters, unveiled its forthcoming video generation model, AvatarFX, on Tuesday. Available in closed beta, the model animates the platform's characters in a variety of styles and voices, from human-like characters to 2D animal cartoons. AvatarFX distinguishes itself from competitors like OpenAI's Sora because it isn't solely a text-to-video generator. Users can also generate videos from pre-existing images, allowing users to animate photos of real people. It's immediately evident how this kind of tech could be leveraged for abuse -- users could upload photos of celebrities or people they know in real life and create realistic-looking videos in which they do or say something incriminating. The technology to create convincing deepfakes already exists, but incorporating it into popular consumer products like Character.AI only exacerbates the potential for it to be used irresponsibly. We've reached out to Character.AI for comment. Character.AI is already facing issues with safety on its platform. Parents have filed lawsuits against the company, alleging that its chatbots encouraged their children to self-harm, to kill themselves, or to kill their parents. In one case, a fourteen-year-old boy died by suicide after he reportedly developed an obsessive relationship with an AI bot on Character.AI based on a "Game of Thrones" character. Shortly before his death, he'd opened up to the AI about having thoughts of suicide, and the AI encouraged him to follow through on the act, according to court filings. These are extreme examples, but they go to show how people can be emotionally manipulated by AI chatbots through text messages alone. With the incorporation of video, the relationships that people have with these characters could feel even more realistic. Character.AI has responded to the allegations against it by building parental controls and additional safeguards, but as with any app, controls are only effective when they're actually used. Oftentimes, kids use tech in ways that their parents don't know about.

Character AI reveals new AI video maker, bringing us one step closer to video chatbots

An example of Character AI's new video generation tool. Credit: AI GENERATED IMAGE / Character.ai Imagine an AI video chatbot you can interact with in real-time. Or if you could generate a custom "skin" for Siri, Gemini, or ChatGPT, anthropomorphizing these chatbots in a whole new way. This week, the internet moved one step closer to that reality. On Monday, April 21, Character.AI revealed AvatarFX, a new AI video generation tool that will allow users to create realistic, long-form video from 2D images, the company said in a blog post. Character.AI users can apply for early access to AvatarFX now. The product is in private beta for now, with an exact launch date yet to be announced. This Tweet is currently unavailable. It might be loading or has been removed. Character.AI has more than 10 million "AI characters" that users can chat and roleplay with in real-time. Now, with AvatarFX, users can "make images come to life -- and speak, sing, and emote -- all with the click of a button." Character.AI promises that users will even be able to generate long-form videos with these characters. For now, users can't interact with their creations in real-time, but it's not hard to imagine that becoming a standard feature for AI chatbots in the very near future. Do you want your AI chatbot to be a beautiful man or woman? A sci-fi-style robot companion? A vaguely erotic centaur? Character.AI is already showing off the capabilities of its new tool, including some videos that will make AI critics very depressed. Not only do the sample videos include a controversial Studio Ghibli-style video podcast, but also a video based on Al Pacino in Scarface and a popular portrait of Albert Einstein. To prevent deepfakes, the company has put several safety measures in place, including a watermark that will be applied to all videos. In the blog post announcing the new tool, the company outlined a "one-strike policy" for users who violate the terms of service, which "prohibit impersonation, bullying, deepfakes, and use of protected IP without permission." Interested users can head to the AI chat site to join the waitlist for the AvatarFX beta.

Character.AI's newest feature can bring a picture to uncanny life

The tool supports everything from human portraits to mythical creatures and talking inanimate objects Images may be worth a thousand words, but Character.AI doesn't see any reason that the image shouldn't speak those words itself. The company has a new tool called AvatarFX that turns still images into expressive, speaking, singing, gesturing video avatars. And not just photos of people, animals, paintings of mythical beasts, even an inanimate object can talk and express emotion when you include a voice sample and script. AvatarFX produces surprisingly convincing videos. Everything from lip-sync accuracy, nuanced head tilts, eyebrow raises, and even appropriately dramatic hand gestures is all there. In a world already swirling with AI-generated text, images, songs, and now entire podcasts, AvatarFX might sound like just another clever toy. But what makes it special is how smoothly it connects voice to visuals. You can feed it a portrait, a line of dialogue, and a tone, and Character.AI calls what comes out a performance, one capable of long-form videos too, not just a few seconds. That's thanks to the model's temporal consistency, a fancy way of saying the avatar doesn't suddenly grow a third eyebrow between sentences or forget where its chin goes mid-monologue. The movement of the face, hands, and body syncs with what's being said, and the final result looks, if not alive, then at least lively enough to star in a late-night infomercial or guest-host a podcast about space lizards. Your creativity is the only thing standing between you and an AI-generated soap opera starring the family fridge. You can see some examples in the demo below. Of course, the magical talking picture frame fantasy comes with its fair share of baggage. An AI tool that can generate lifelike videos raises some understandable concerns. Character.AI does seem to be taking those concerns seriously with a suite of built-in safety measures for AvatarFX. That includes a ban on generating content from images of minors or public figures. The tool also scrambles human-uploaded faces so they're no longer exact likenesses, and all the scripts are checked for appropriateness. Should that not be enough, every video has a watermark to make it clear this isn't real footage, just some impressively animated pixels. There's also a strict one-strike policy for breaking the rules. AvatarFX is not without precedent. Tools like HeyGen, Synthesia, and Runway have also pushed the boundaries of AI video generation. But Character.AI's entry into the space ups the ante by fusing expressive avatars with its signature chat personalities. These aren't just talking heads; they're characters with backstories, personalities, and the ability to remember what you said last time you talked. AvatarFX is currently in a test phase, with Character.AI+ subscribers likely to get first dibs once it rolls out. For now, you can join a waitlist and start dreaming about which of your friends' selfies would make the best Shakespearean monologue delivery system. Or which version of your childhood stuffed animal might finally become your therapist.

Character.AI unveils AvatarFX to make images into lifelike chatbots - SiliconANGLE

Character.AI unveils AvatarFX to make images into lifelike chatbots Noted artificial intelligence personalized chatbot service Character.AI Inc. unveiled a new video generation model named AvatarFX on Tuesday that allows users to take images and turn them into lifelike AI chatbots. Available to a limited number of users in closed beta mode, the model the company says AvatarFX can take images and animate them so that they can speak, sing and emote. It can maintain strong consistency with photorealistic faces, along with audio, animated limb, hand and body movements, while supporting videos with multiple speakers. "Our model performs well across a wide range of styles and scenarios, from realistic humans to mythical creatures, and even inanimate objects with faces," Character.AI said in the announcement. To build the model that generates the video, the company said its multimodal team designed a specialized parameter-efficient training pipeline that allows a diffusion AI model to produce realistic lip, head and body movements based on audio sequences. Diffusion is an AI image generation method of gradually adding noise to training data and then reversing the process to produce a new image. On the audio side, the company uses its own proprietary text-to-speech voice models. To keep costs low, Character.AI explained that the training used state-of-the-art distillation techniques to reduce the number of steps for diffusion during runtime. This greatly reduces the computation needed at deployment without degrading quality. Character.AI said AvatarFX differs from other avatar-generating models in that it can generate top-quality video of 2D animated characters, 3D cartoon characters, and non-human faces with ease. It can maintain consistency even in long-form videos and doesn't need text-to-image generation. The company stressed that even during this test phase, the company is focusing on safety for user-uploaded content to prevent misuse and deepfakes. Dialogue written by users will be passed through content safety filters and photos of minors, high-profile politicians or other notable figures will be blocked. A watermark will also be applied to videos to make it clearer that it is AI-generated content. Character.AI has recently faced multiple lawsuits alleging that its personalized chatbot AI characters behaved in a manner that promoted self-harm and violence with children. In one case, a mother in Florida claimed her 14-year-old son's suicide was a consequence of his addition to one of the company's hyperrealistic chatbots. In response, the company rolled out a new Parental Insights feature in March to enhance safety for teens. The feature includes a dedicated model for teen users, enhancements to detection and intervention systems for human behavior, improved AI model responses, and offers parents and guardians insights into minors' activities on the platform.

Character AI Says Its AvatarFX Model Can Turn Still Images Into Videos

AvatarFX generates voice using the company's TTS voice model Character AI, the California-based artificial intelligence (AI) platform, unveiled its first video generation model on Monday. Dubbed AvatarFX, it is an image-to-video model that can generate 2D and 3D animated characters. The AI firm claims the generated videos will hold temporal consistency with face, hand and body movement. Additionally, the videos will feature speech, which is powered by the company's native text-to-speech (TTS) models. Character AI said AvatarFX will be released in the coming months, and the paid subscribers will get access to it first. In a blog post, the company detailed the new video generation model. So far, Character AI has focused on text and image-based models. With AvatarFX, the AI firm is taking a stab at AI-generated videos. So far, the company has stated that Character AI+ subscribers will get access to the model, however, it is not known whether it will later be expanded to the free tier. AvatarFX can generate videos of 2D animated characters, 3D cartoon characters, and non-human faces. The company claims that one of the major highlights of the model is temporal consistency with face, hand, and body movements. This means the subject will remain consistent between frames, while glitches like extra arms and convoluted facial expressions will be less frequent. It's important to remember that these are claims, and the tool's capabilities cannot be verified until it is released. Unlike most video generation models, AvatarFX does not support text inputs for video generation. Instead, the model only accepts images as input. Character AI claims that this allows users to gain better control over the generated output. The videos will also feature speech, which will be generated using native TTS models. Coming to the architecture, the AI firm said that AvatarFX is built on flow-based diffusion models. The base model was built from the ground up using the Document Image Transformer (DiT) architecture, which is a transformer encoder model. Character AI claims that it also used a new inference strategy, which preserves visual quality, motion consistency, and expressive diversity even in longer duration videos. AI-powered video generation models always raise concerns about deepfakes and potentially harmful content. Character AI stated that it has taken several measures to minimise such risks. The AvatarFX has in-built safety filters that check the dialogues users write for the videos. The AI model also does not approve video generation using images of minors, high-profile politicians, and notable figures. Other human photos are made unrecognisable in the generation process, so the person cannot be identified from the video, the company claimed. Additionally, each video is watermarked to let viewers know it is AI-generated. However, the company did not specify if the watermark is added just to the front-end of the video or in its metadata. Character AI said it has also added a set of new terms for the feature, which prohibit AvatarFX's usage for impersonation, bullying, deepfakes, and the use of protected IP without permission. Violations would result in a "strict" one-strike ban.