We’ve raised $1.3M from YC & Afore Capital so you never have to type or speak again. Typing is slow. Talking into your computer is awkward. Subvocal fixes both Our glasses give you the speed of speech without you making a sound or touching a keyboard. It tracks your internal speech muscles with non-contact sensing and decodes silent, almost invisible mouthing into text at more than 200 words/min. Pre-orders are now live at:

Introducing RL with Neural Feedback (RLNF). We have found reliable ways to train AI models to respond to human preferences purely from brain signals. The example below is a picture of a face that self-transforms into a more attractive version for the subject as they are looking at it.

An astronaut works to repair a damaged communication relay on a frozen moon, surrounded by an incredible contrast of icy blue planetary light and a glowing amber work lamp. With Flova 1.5, Sci-Fi Live-Action Films brings the astronaut and environment to life, while Cinematic Lighting Director adds depth, atmosphere, and cinematic lighting. You can combine multiple Skills within one project to enhance different creative elements while keeping the same scene and references intact. Created with Flova @Flovaai Explore your own cinematic AI workflow

Flova 1.5 | A New Creative Experience Let Agent organize your assets so you can focus on creating! - One Project. Three Views. Switch seamlessly between List, Grid, and Canvas. - Flexible Asset Organization. Combine and nest Groups and Sequences to organize assets your way. - Smart Search & Filters. Find exactly what you need, fast. 🚀Stay organized. Let ideas flow!

Halloween takes a terrifying turn when Michael Myers gets creative with a pumpkin! 🎃🔪 @VivixLabs_HQ Vivix W1 brings this spooky transformation to life with striking visual details, eerie lighting, and smooth, natural motion. The way the scene maintains its creepy atmosphere throughout the transformation is seriously impressive, especially with such fast generation and low costs. I’m posting this to enter the Vivix Creator Contest! Ready to put your creativity to the test? Create something unforgettable and join the challenge too. 👻🔥

Who did this? 🤣 A game where you can cosplay as @ElonMusk or Ambani If you're Ambani, there is a surprise waiting for you Make your guy win 💀 Jio mere laal

That Baddie Again 🥵 Say NO to Hollywood and YES to AI. Seedance 2.5 on @frameo_ai #MadeWithFrameo# Prompt : Create a 30-second, 1080p ultra-realistic personal home-video showing an ordinary summer afternoon in the life of a young Korean woman. No reference image. **MAIN SUBJECT** A naturally beautiful Korean woman in her early 20s, with realistic skin texture, minimal makeup, and a relaxed, confident personality. Long black hair loosely tied into a messy ponytail, with a few strands framing her face. She wears a fitted pastel-blue crop top, loose cream-colored casual trousers, black sneakers, and a simple silver necklace. Maintain her exact facial features, hairstyle, outfit, body proportions, and appearance throughout the entire video. **SETTING** A quiet older residential neighborhood in Seoul on a warm summer afternoon. Narrow concrete lanes, small houses, weathered walls, potted plants, parked bicycles, overhead utility wires, laundry hanging outside homes, a tiny convenience store, and children playing football. Everything feels lived-in, peaceful, and authentic, with no tourist attractions or recognizable commercial brands. **CAMERA / VISUAL AESTHETIC** Raw personal footage casually recorded by a friend using an early-2000s consumer DV camcorder. Strong handheld shake, imperfect framing, autofocus hunting, exposure shifts, faded colors, soft digital detail, mild image noise, occasional motion blur, accidental zooms, and natural camera imperfections. No stabilization, drone shots, gimbal movement, dramatic lighting, or polished commercial cinematography. **SCENE 1 — OUTSIDE THE HOUSE** She steps outside carrying a small reusable shopping bag, locks the door, adjusts her messy ponytail, and gives the camera a playful smile. She starts walking down the narrow residential lane as a light summer breeze moves her hair. **SCENE 2 — MEETING THE SCHOOLCHILDREN** Several schoolchildren with backpacks approach from the opposite direction. Recognizing them, she raises her hand and smiles. Woman: “Hey, aren't you guys late?” The children laugh and continue walking. One turns around to wave at her before catching up with the others. She laughs naturally and continues down the street. **SCENE 3 — A QUICK FOOTBALL GAME** A little farther along, neighborhood children are playing football in the street. The ball rolls toward her and stops beside her sneakers. She looks down at it, then toward the children. One child calls out: “Noona! Pass it back!” She grins, takes a small step, and gently kicks the ball toward them. It rolls naturally across the concrete, and the children chase after it excitedly. She laughs at their enthusiasm and resumes walking. The ball stays with the children without disappearing or duplicating. **SCENE 4 — A LITTLE SUMMER MOMENT** She pauses beside a small outdoor water tap, rinses her hands, and shakes off the droplets. A bicycle passes behind her as she fixes a loose strand of hair and gives the camera a teasing, confident look. She says: “It's way too hot today!” The camera operator chuckles softly behind the camera. She smiles and keeps walking. **SCENE 5 — THE GOODBYE** At the corner of the neighborhood, she turns to watch the children playing football farther down the lane. One notices the camera and waves enthusiastically. She waves back, then turns toward the camera with a small playful smile. Woman: “See you tomorrow!” She walks around the corner while the camera follows her for a few steps. The footage ends with an abrupt cut to black, like an authentic home-video recording. **AUDIO** Natural location sound only: children's laughter and shouting, football bouncing against concrete, footsteps, birds, summer insects, bicycle bells, distant scooters, neighborhood conversations, leaves rustling in the breeze, water splashing, and subtle camera-handling noise. No background music, no narration, and no additional dialogue beyond the specified lines.

Sometimes, AI-generated product ads just don’t meet the client’s expectations for visual clarity. @HitPawofficial HitPaw VikPea makes it easy to upscale them to 4K for sharper details and a more professional look. Try HitPaw VikPea here >>

He can buy his own 😂 Made using Seedance 2.5 Prompt : Create an ultra-realistic live-action revenge-comedy scene on the outdoor steps of a busy Seoul shopping street. Natural handheld camera movement, realistic physical interactions, authentic street ambience, expressive facial reactions, and fast-paced Korean action-comedy energy. No animation or video-game visuals. CHARACTERS: A stylish Korean woman in her 20s with long black hair, wearing a fitted crop top, casual denim shorts, and sneakers. She sits on concrete steps enjoying tacos from a takeaway tray, with several sauce packets and a drink beside her. A young Korean man in a black T-shirt, loose cargo pants, and sneakers approaches her. ACTION SEQUENCE: The woman sits comfortably on the steps, taking a bite of her taco and enjoying her food while scrolling on her phone. The man walks past, stops beside her, and makes a rude comment. He deliberately snatches a taco from her tray, takes a bite, and laughs mockingly when she protests. He then squeezes sauce from a packet onto her shorts and smirks at her, expecting her to cry. She slowly puts down her taco, looks at the stain, then turns toward him with an icy expression. Without warning, she grabs a sauce packet, squeezes a generous streak of red taco sauce across his cheek and nose, and gives him a sarcastic smile. He freezes in disbelief as nearby pedestrians turn to look. Enraged, he steps toward her. She stands up, sidesteps his clumsy advance, and gives him a quick, exaggerated shove that sends him stumbling backward into a large open rubbish bin beside the stairs. The bin rocks violently as he lands inside among empty takeaway containers. Keep the action comedic, non-graphic, and physically believable. She calmly picks up her taco tray, brushes off her shorts, and looks down at him. She says, “Next time, buy your own tacos.” She takes another bite and walks away confidently while he struggles to climb out of the bin, his face still covered in sauce. SOUND DESIGN: Busy Seoul street ambience, footsteps, taco-wrapper crinkles, a bite of food, mocking laughter, sauce packet squeezing, startled reactions, a loud bin rattle, and the woman's clear dialogue. No overpowering background music. VISUAL REQUIREMENTS: Maintain consistent faces, outfits, props, and location. Clearly show the man's misbehavior before her revenge. Show the sauce visibly landing on his face before the comedic bin incident. Keep the action readable in continuous handheld shots, with realistic movement, natural facial expressions, accurate object interactions, and a satisfying final reaction. NEGATIVE PROMPT: No blood, serious injuries, supernatural powers, cartoon physics, distorted anatomy, face drift, disappearing props, unrealistic food interactions, excessive slow motion, subtitles, logos, or watermarks.

Computer activity, stored over time, becomes human traces: a record of the actions, decisions, and context behind knowledge work. Today we’re releasing HelixCove, a dataset of work unfolding across an organization.

Cotton Candy Baddie 🍭 | Made using Seedance 2.5 Prompt: Create a photorealistic 30-second personal vlog featuring the same stylish young Korean woman in her early 20s enjoying a playful afternoon on the streets of Seoul. CHARACTER: Long black hair in a loose ponytail, realistic skin texture, glossy lips, subtle makeup, fitted baby-pink top, low-rise jeans, silver jewelry, and black sneakers. Maintain consistent facial identity and outfit. STORY: She walks confidently down a colorful shopping street, holding a giant pink cotton candy. She takes a bite, notices the sugar sticking to her lips, and gives the camera a cheeky smirk. She spins the cotton candy playfully, accidentally gets a little on her fingers, and laughs while trying to clean them. Passing a street vendor, she buys a blue cotton candy, holds both colors beside her face, and poses like a playful baddie. She ends by walking away while looking back at the camera with a confident smile. VISUAL STYLE: Early-2000s MiniDV footage, 4:3 aspect ratio, handheld camera shake, imperfect framing, faded colors, natural sunlight, subtle grain, authentic candid moments. AUDIO: Street ambience, footsteps, vendor chatter, wrapper sounds, and natural laughter. No music or narration. NEGATIVE PROMPT: No face drift, distorted hands, unnatural expressions, CGI, anime, beauty filters, subtitles, logos, or watermarks.

Kling 4.0 feels like a major leap for AI filmmaking. @Kling_ai The fluidity of the movement, finer control over each scene, and richer visual storytelling are pushing generated video into a whole new space. Really curious to see what creators build with it next.

Kling 4.0 is launching this October, and it’s bringing some serious upgrades. 🔥 4K + 10-bit HDR, better lip-sync and stereo audio, up to 15 references, 10-keyframe control, and native 30-second generation. Kling 4.0 Flash is already live, so creators can get an early taste of what’s coming.

HAMOD — the hair dryer that makes every styling session effortless WAN 3.0 of @alibaba_cloud on @openart_ai #OpenArtAdAwards# Reference Prompt : Create a ultra-realistic UGC -style beauty video featuring a young woman casually reviewing a premium hair dryer called HAMOD. She is filming herself at home with a smartphone, creating an authentic TikTok/Reels-style product review. She wears white cozy pajama bottoms and a soft pastel-blue pajama top, creating a relaxed, cozy aesthetic.

Smartwatch Commercial — Ready for Every Moment Seedance 2.5 of @BytePlusGlobal On @openart_ai #OpenArtAdAwards# Prompt : Create a 30-second, 1080p ultra-realistic cinematic commercial for a fictional premium smartwatch brand called VANTA. The film should feel like a high-end global technology and sports campaign, showing the watch performing across completely different environments. A stylish young woman with long dark hair wears the same VANTA smartwatch throughout the film. Keep the watch design, color, proportions and display consistent. Opening — City: She walks through a modern city wearing a black fitted athletic top, charcoal leggings and running shoes. The watch display shows heart rate and activity tracking as she starts running. Swimming: Cut seamlessly to her in a swimming pool wearing a sleek black athletic swimsuit. Underwater shots show her swimming while the smartwatch remains clearly visible, tracking her workout. Rain: She emerges outdoors wearing a black waterproof running jacket, black leggings and running shoes as heavy rain begins. She continues running through wet streets while water runs across the watch. Show a close-up of the display remaining active. Mountain: Transition to a mountain trail. She wears a dark green performance jacket, black hiking leggings and trail shoes. She climbs while the watch displays heart rate, distance and elevation. Gym: She is now wearing a dark sports bra, black leggings and training shoes, performing strength exercises. Brief close-ups show workout tracking and heart-rate data. Final — Sunset: She stands on a rooftop wearing a minimal black athletic jacket and leggings, watching the sunset. The camera moves from her face to the smartwatch for a detailed hero close-up. Voiceover throughout: “Built for every environment. Track your heart rate, your workouts, your progress and every step forward. From the pool to the pavement, from rain to the highest trail—VANTA keeps up with you. One watch. Every move.” Style: Photorealistic, cinematic sports cinematography, realistic water and rain physics, detailed smartwatch materials, dynamic tracking shots, macro product shots, natural movement, dramatic lighting, shallow depth of field and premium technology advertising. Negative: No watch design changes, inconsistent display, identity drift, outfit glitches, distorted hands, extra limbs, duplicated people, fake-looking water, excessive CGI, random text, subtitles or watermark.

When the world gets loud, ECHO lets you hear what matters. 🎧 Made on @openart_ai using @BytePlusGlobal Seedance 2.5 model #OpenArtAdAwards# Prompt : Create a 30-second, 1080p ultra-realistic cinematic commercial for fictional premium wireless headphones ECHO. A stylish young woman with long dark hair, natural makeup, black fitted top, oversized neutral jacket, wide-leg trousers and sneakers walks through a noisy modern city. Maintain consistent character, outfit and headphone design. She takes out the ECHO headphones and puts them on. The chaotic city noise suddenly fades, replaced by immersive music. As she walks, the environment subtly reacts to the music—traffic lights pulse, city lights glow rhythmically and people move naturally with the beat. She enters a subway where footsteps, train sounds and commuters blend into the music. Show her enjoying the experience with wide shots, tracking shots, facial close-ups and occasional detailed headphone shots. She exits into golden-hour light and walks along a rooftop, adjusting the headphone controls. End with her overlooking the city, eyes closed, completely immersed, followed by a cinematic hero close-up of the headphones. Voiceover: “Some sounds demand your attention. Others deserve to disappear. ECHO brings you closer to what matters, with immersive sound, powerful clarity and lasting comfort. Block out the noise. Feel every detail. Find your own frequency.” Style: Premium technology campaign, photorealistic, cinematic lighting, realistic skin and hair, smooth camera movement, shallow depth of field, natural city ambience, rich music and sophisticated sound design. Negative: No headphone morphing, identity drift, outfit changes, distorted anatomy, duplicated people, excessive CGI, subtitles, captions, random text or watermark.

Baddie on a weekend Made using Seedance 2.5 Prompt : Create a 30-second, 1080p ultra-realistic early-2000s DV home video of a young Korean woman spending a relaxed Saturday afternoon in an older Seoul neighborhood. Young Korean woman, early 20s, natural beauty, realistic skin, minimal makeup, black wavy hair in a messy side ponytail, fitted pastel-blue short top, loose cream pajama-style pants, black sneakers and simple silver necklace. Keep her identity, outfit, hairstyle and proportions perfectly consistent throughout. She leaves her house with a cheerful expression and walks through a quiet residential lane carrying a small reusable shopping bag. She notices a few school children struggling to get their kite down from a low tree branch. She walks over, notices a ladder leaning against the tree, carefully climbs the ladder, reaches the kite and gently frees it from the branches. She climbs down and happily hands the kite back to the children. The kids smile and thank her before running off. She continues walking and notices a friendly cat sitting beside the street. She stops, crouches down and gently pats the cat, smiling as it brushes against her hand. She then continues down the lane. A bicycle cart seller passes nearby selling colorful lollipops. She stops, chooses a lollipop, gives the seller a few coins, receives the candy, unwraps it and puts it in her mouth. She continues walking home while happily enjoying the lollipop. A gentle breeze suddenly moves through the street, lifting her hair and the leaves around her. She smiles, adjusts her messy side ponytail with one hand and keeps walking. Near the end, she turns toward the camera with the lollipop still in her mouth, smiles naturally and says, “Happy Saturday!” She waves lightly and continues walking home. Camera: Raw handheld early-2000s DV camcorder, imperfect framing, autofocus hunting, faded colors, exposure shifts, soft digital detail, mild digital noise, natural motion blur and occasional awkward zooms. No polished cinematography. Audio: Natural neighborhood ambience only — footsteps, children talking, kite movement, ladder creaks, cat sounds, bicycle-cart bell, coins, street sounds, birds, wind and distant traffic. No music, no narration, no subtitles. Continuity: One woman throughout, consistent face, hairstyle, outfit and proportions. Kite remains with the children after she gives it to them. Lollipop remains in her hand/mouth after purchase. Shopping bag stays with her. Realistic ladder climbing, hand movements and object interactions. Negative: CGI look, plastic skin, identity drift, outfit changes, duplicate people, teleporting objects, distorted hands, extra fingers, floating objects, unrealistic physics, random characters appearing, subtitles, logos, watermark.

These shoes are built to keep you moving, wherever the road takes you. 👟 Made using @BytePlusGlobal Seedance 2.5 model on @openart_ai #OpenArtAdAwards# Prompt : Create a 1080p ultra-realistic cinematic sportswear commercial featuring a confident young woman wearing black sports leggings, a fitted athletic top and premium running shoes. Keep the girl visible throughout the video, with the shoes naturally integrated into the story rather than showing only product close-ups. She walks through a modern city before starting to run. Use wide shots showing her full body and face, side tracking shots, front-facing shots and occasional low-angle shots highlighting the shoes. She reaches a park, stretches on a bench, adjusts her shoes and continues walking before picking up her pace again. Show her lifestyle, movement and confidence, with brief close-ups revealing the shoe’s cushioning, flexibility, lightweight construction and detailed materials. Rain begins. She walks through the wet streets, then runs through the rain. Capture her full body, facial expressions, reflections and natural movement, mixed with a few slow-motion shots of the shoes gripping the wet pavement and creating realistic water splashes. As the rain fades, she slows down and confidently walks forward. End with a cinematic shot showing the girl and shoes together, followed by a tasteful close-up of the shoes on the wet reflective pavement. Voiceover throughout: “Every step starts with the right foundation. Lightweight, flexible and designed for comfort. Responsive cushioning supports every stride, while reliable traction keeps you moving when the road gets wet. Built for every mile, every moment, every move. Shoes made to keep you moving.” Style: Premium global sportswear campaign, photorealistic skin and fabrics, realistic athletic movement, cinematic lighting, rain and reflections, natural motion blur, shallow depth of field, dynamic tracking, subtle slow motion and high-end commercial cinematography. Negative: No shoe-only video, no missing girl, no shoe design changes, no identity drift, no distorted feet, extra limbs, duplicated people, outfit changes, text, subtitles or watermark.