AI Educator & Creator 🧩 Teaching Generative AI, Creator Tools & Emerging Tech | 1M+ followers - (hii I’m new to X)

London, England
Hot take: Kling 3.0 quietly killed the entire β€˜talking head’ AI workflow.πŸ”₯ We just stress-tested Kling 3.0 and the lip sync is ridiculous. Ran a Deadpool scene + a few celebrity-style tests (parody) and it kept character identity locked across cuts! Check this outπŸ‘‡πŸ‘‡πŸ‘‡
40
73
882
72,947
Grok can make insane UGC visuals now.😱 This looks like a real coffee ad shoot… but it’s AI.πŸ˜€ Prompt below.πŸ‘‡
3
3
245
Prompt: Create a highly realistic UGC-style product photo of a young woman outdoors on a cozy balcony during golden hour. She has long dark brown wavy/curly hair, natural makeup, warm tan skin, and a bright happy expression. She is wearing a white ribbed tank top and is framed from the waist/chest up in a vertical portrait composition. She is holding a bright blue coffee pouch in her left hand, facing the camera clearly. The pouch should have bold white text that reads β€œTHE GOAT” in large capital letters. Near the bottom of the pouch, include a small white goat icon and the text β€œcoffee with a kick” underneath it. In her right hand, she is holding a clear glass of iced coffee / iced latte with visible ice cubes, creamy swirling milk mixing into the coffee, and a silver metal straw. The drink should look refreshing and visually appealing. Her facial expression should feel natural, candid, and cheerful, with her mouth slightly open in a smile as if reacting to someone off camera. She should be looking slightly to the side, not directly into the lens. The background should show a small balcony or terrace with black decorative metal railings and several terracotta plant pots filled with green herbs/plants. The background should feel warm, lifestyle-driven, and authentic, with a soft urban residential feel. Style should be: Ultra realistic Natural lighting UGC / influencer ad aesthetic Warm golden hour sunlight Soft shadows Shallow depth of field Lifestyle brand content Casual, premium, organic coffee ad vibe Camera / composition notes: Vertical 9:16 Medium close-up Sharp focus on the woman, coffee pouch, and drink Slight background blur Natural skin texture Clean hands and fingers Product fully visible and readable Photorealistic, not overly polished or studio-looking Optional quality tags: photorealistic, UGC ad, influencer content, golden hour, realistic skin texture, premium coffee branding, balcony lifestyle, high detail, natural composition, commercial product photography
1
66
Sebastien Jefferies retweeted
VR finally stopped looking like a giant headset.🚨 Meta just unveiled its new VR Glasses! β†’ Around 100g β†’ Eye + hand tracking β†’ Full spatial computing β†’ Compute moved into an external puck β†’ Much smaller, lighter form factor β†’ Built to run the same ecosystem as Meta Quest This is the kind of VR hardware we’ve all been waiting forπŸ”₯
1
4
272
AI filmmaking is getting dangerously close to changing the film industry forever. 🀯  We made all of this using Seedance 2.5 + simple prompts. It’s becoming less about typing a prompt and hoping for something cool… And more about actually directing the result. If you want to learn how we create AI films, ads, characters and cinematic content like this, join us inside AI Creator Lab β†’ lnkd.in/ekkpXGT3
3
1
8
627
Grok animations are getting serious πŸ”₯ Prompt πŸ‘‡
1
9
734
Create an EXACTLY 15-second fully animated cinematic boxing sequence. 16:9. 4K. 24fps. SOUND ON. FULLY ANIMATED FROM FIRST FRAME TO LAST FRAME. NO LIVE-ACTION FOOTAGE. NO PHOTOREALISTIC HUMAN. NO FIGHTING. NO PUNCHES. NO DIALOGUE. NO TEXT. VISUAL STYLE: Premium theatrical hand-drawn 2D animation. Mature dramatic sports-film aesthetic. Beautiful hand-inked linework. Expressive hand-drawn faces. Detailed cel shading. Painted backgrounds. Rich atmospheric lighting. Strong cinematic composition. Natural human anatomy. Fluid character animation. Subtle frame-to-frame line variation. Hand-painted shadows and highlights. Beautifully animated cloth, sweat, hair and smoke. The result should feel like an expensive feature-length animated boxing drama, NOT a children's cartoon. The boxer should look fully illustrated and animated in EVERY frame. No live-action textures. No photorealistic face. No 3D-rendered character. No realistic footage with an animation filter. ──────────────────────── THE BOXER ──────────────────────── One adult male professional boxer, around 30. Lean athletic fighter physique. Short messy dark hair. Short dark stubble. Defined but believable face. Slight old scar cutting through one eyebrow. Tired eyes. Subtle shadows underneath the eyes. Strong nose. Natural lips. Slightly weathered appearance. His face should have CHARACTER rather than perfect beauty. Black satin boxing robe. White hand wraps. Deep burgundy-red boxing gloves. Black boxing shorts. Black boxing boots. His design must remain perfectly consistent throughout the entire animation. Same: face, hair, scar, body, robe, gloves, proportions. ──────────────────────── FACIAL PERFORMANCE β€” MOST IMPORTANT ──────────────────────── The face is the emotional center of the animation. DO NOT use exaggerated anime expressions. DO NOT make him constantly angry. DO NOT make him scream. His emotions should be communicated through beautifully animated MICRO-EXPRESSIONS. Subtle eye movement. Natural blinking. Small eyelid tension. Tiny eyebrow movement. Controlled breathing. Nostril movement. Small jaw tension. Lips pressing together briefly. A swallow. Tiny movement in his cheek. Small changes in where his eyes focus. The audience should understand: he is nervous, he knows what is coming, he is controlling the adrenaline, he is mentally preparing, he gradually becomes completely focused. His expression evolves throughout the 15 seconds. Not through one giant facial change. Through dozens of tiny movements. ──────────────────────── 0:00–0:04 β€” DARKNESS ──────────────────────── OPEN IN NEAR BLACK. A narrow concrete arena tunnel. The boxer slowly walks toward a distant rectangle of brilliant white arena light. Camera tracks backward in front of him. MEDIUM CLOSE-UP. Upper chest and face. The background is dark and painterly. Overhead fluorescent lights pass above him one at a time. LIGHT. SHADOW. LIGHT. SHADOW. Each light briefly reveals different parts of his animated face. First: his eyes. Next: his nose and cheek. Next: his entire face. His robe gently moves with each step. Hair shifts subtly. He breathes through his nose. Tiny nostril movement. His eyes remain fixed ahead. A small eye movement toward the bright tunnel exit. One slow natural blink. His jaw tightens slightly. Then releases. The animation is restrained and mature. ──────────────────────── 0:04–0:07 β€” INTO THE LIGHT ──────────────────────── The arena entrance grows larger behind camera. Warm-white light gradually reaches his face. Camera moves closer. TIGHT THREE-QUARTER PORTRAIT. Beautiful hand-drawn facial detail. Fine ink lines around eyes. Individual illustrated stubble marks. Tiny beads of animated sweat. One bead forms near his temple. It slowly travels downward. His lips press together. Release. He swallows once. His throat moves subtly. Thenβ€” he steps through the tunnel. WHOOOOOOOOSHβ€” the enormous arena opens around him. Crowd sound suddenly expands. His eyes react before the rest of his face. Tiny widening. Not shock. Adrenaline. Then they narrow slightly again. Focus returning. ──────────────────────── 0:07–0:10 β€” GLOVES / ARENA ──────────────────────── Camera smoothly lowers from his face toward his hands while he continues walking. CLOSE ANIMATED DETAIL: burgundy boxing gloves. Beautiful hand-drawn leather texture. Creases. Stitching. Highlights sliding across the curved surface. His fingers tighten inside one glove. The leather compresses. CREEEAK. His arm swings naturally with his walking rhythm. Behind him: the enormous animated arena becomes visible. Thousands of spectators represented through layered hand-drawn silhouettes. Camera flashes flicker throughout the darkness. A gigantic boxing ring glows beneath four enormous white spotlights. Volumetric shafts of illustrated light cut through atmospheric haze. Camera begins rising again. Gloves β†’ chest β†’ neck β†’ jaw β†’ FACE. ──────────────────────── 0:10–0:13 β€” THE FACE ──────────────────────── Now the most cinematic portrait of the sequence. TIGHT THREE-QUARTER CLOSE-UP. He continues walking. The background becomes a beautifully painted blur of: arena lights, crowd, camera flashes, haze, boxing ring. One side of his face is illuminated. The other disappears into rich painted shadow. A camera flash fires. For ONE FRAME: his entire face becomes brightly illuminated. Every drawn detail becomes visible. Sweat. Stubble. Scar. Eyes. Skin shading. Then immediatelyβ€” shadow again. He takes one deeper breath. Chest subtly rises. His eyes move toward the ring. Jaw tightens. A bead of sweat reaches his cheek. He blinks once. When his eyes reopenβ€” something has changed. The nervousness is gone. Completely focused. ──────────────────────── 0:13–0:15 β€” THE RING ──────────────────────── He reaches the boxing ring. Camera moves slightly lower. Foreground ring ropes create strong graphic horizontal lines across the composition. He places one burgundy glove against the top rope. The rope bends naturally. Trainer's hands briefly separate the ropes. The boxer steps through. Camera remains CLOSE. He enters the ring. Then stops. Camera slowly rises back toward his face. Arena lights blaze above him. Crowd becomes enormous. He lifts his eyes toward the opposite corner. No smile. No anger. No fear. Just complete concentration. One tiny controlled exhale. His glove slowly closes into a fist. LEATHER CREAK. His eyes remain perfectly still. CUT TO BLACK. EXACTLY 15.00 SECONDS. ──────────────────────── ANIMATION STYLE ──────────────────────── PREMIUM HAND-DRAWN 2D FEATURE ANIMATION. Traditional frame-by-frame character animation. Organic ink outlines. Slight natural line variation between frames. Sophisticated adult character design. Realistic human anatomy translated into illustration. Detailed cel shading. Multi-layer painted shadows. Painterly background environments. Hand-painted atmospheric haze. Beautiful illustrated practical lighting. Subtle film-grain texture over finished animation. Animation primarily on twos with smoother animation during: eye movements, blinks, camera movement, walking, cloth movement, sweat, ring-rope movement. The animation should feel intentionally DRAWN. Do not attempt to hide the medium. ──────────────────────── FACE ANIMATION ──────────────────────── Prioritize facial animation above everything else. Eyes must feel alive. Pupils make tiny adjustments. Eyelids naturally compress around eyes. Blinking should include: upper eyelid movement, tiny lower-lid reaction, subtle cheek response. Jaw tension should subtly alter cheek shape. Breathing subtly moves: nostrils, upper lip, neck, shoulders. Sweat follows facial contours. Do NOT animate every facial feature simultaneously. Most of the face remains still. One tiny movement happens. Then another. Then stillness. This restraint makes the performance powerful. ──────────────────────── LIGHTING ──────────────────────── Dramatic animated chiaroscuro lighting. Tunnel: deep navy-black shadows, cold overhead fluorescent lights, bright white arena opening. Arena: powerful white stadium spotlights, warm skin highlights, deep cool shadows, occasional camera flashes, visible beams through atmospheric haze. Light should move across the character as he walks. Shadows change according to his physical position. Do not keep his face uniformly illuminated. Use darkness aggressively. ──────────────────────── CAMERA ──────────────────────── Cinematic animated camera choreography: MEDIUM CLOSE-UP TUNNEL TRACK β†’ TIGHTER THREE-QUARTER FACE β†’ DESCEND TO GLOVES β†’ RISE BACK TO FACE β†’ LOW RING-ROPE ANGLE β†’ FINAL FACE. Camera movement should remain restrained. No wild anime camera. No spinning. No impossible orbit. No rapid zoom. No speed ramp. The camera should feel like a cinematographer filming an animated drama. ──────────────────────── SOUND DESIGN ──────────────────────── NO DIALOGUE. Start almost quiet. Tunnel: BOOM... BOOM... BOOM... Boxing boots against concrete. Robe fabric moving. Quiet breathing. Muffled crowd behind walls. As he approaches the arena: crowd gradually grows. Then when he exitsβ€” ROOOOOOOAR. Huge stadium reverb. Thousands of indistinct voices. Camera shutters. Distant whistles. Low arena rumble. During glove close-up: leather creaking. During facial close-up: bring his controlled breathing slightly forward in the mix. At the ring: rope tension. Canvas footsteps. Final moment: crowd roaring. One controlled exhale. Glove: CREEEAK. BLACK. Optional restrained cinematic score: deep cello, low percussion, subtle bass pulse. Never overpower the environmental sound. ──────────────────────── AVOID ──────────────────────── NO live action. NO photorealistic human. NO rotoscoped realistic footage. NO animation filter over live action. NO 3D CGI. NO Pixar-like 3D. NO glossy rendering. NO plastic skin. NO videogame aesthetic. NO childish cartoon. NO chibi. NO giant eyes. NO exaggerated anime facial expressions. NO screaming. NO glowing eyes. NO magical effects. NO energy aura. NO supernatural powers. NO fighting. NO punches. NO blood. NO opponent attacking. NO violence. NO static face. NO dead eyes. NO frozen expression. NO lip jitter. NO random eyebrow movement. NO exaggerated anger. NO bright colorful arena. NO neon cyberpunk palette. NO music-video aesthetic. NO text. NO subtitles. NO logos. NO watermark. ──────────────────────── FINAL TARGET ──────────────────────── Create fifteen seconds that feel like the opening of a prestigious HAND-DRAWN ANIMATED BOXING FILM. The spectacle isn't a fight. It's the boxer himself. A man walking out of darkness. Thousands of people roaring somewhere ahead. His gloves moving beside him. Arena lights slowly revealing his face. Sweat running down his cheek. His eyes showing tiny traces of nerves. Thenβ€” one breath. one blink. and his entire emotional state quietly changes. He's ready. Beautiful hand-drawn animation. Expressive eyes. Subtle human acting. Dramatic shadows. Massive arena atmosphere. Detailed gloves. Painterly lighting. Mature cinematic storytelling. FULLY ANIMATED. 4K. EXACTLY 15 SECONDS.
293
I’m heading to San Francisco for Meta Connect 2026. 🚨 This year Meta is going much bigger than just social media. We’re talking about the technologies that could completely change how we create, communicate. I’ll be attending as an AI creator and sharing as much as I can from San Francisco. And we’ll be dropping even more exclusive content, breakdowns and behind-the-scenes insights from the event inside AI Creator Lab. Join us here : lnkd.in/ekkpXGT3
7
468
One of the best AI creator communities you can join just opened a 72-hour flash sale. πŸ‘€ Inside AI Creator Lab , we teach you how to actually create with AI, not just play around with new tools. β†’ AI filmmaking β†’ Viral short-form content β†’ AI ads & UGC β†’ Prompting, editing & creative direction For the next 72 hours,we opened just 15 VIP spots for: $240 for the entire year 65% off. If you want one of the last 4 spots, join below ‡️ skool.com/aicreatorlab/about πŸš€
1
4
531
Grok 4K prompt πŸ”₯πŸ‘‡
1
10
644
Create an EXACTLY 15-second photorealistic 4K nature-documentary shot of a real honeybee flying naturally through a beautiful wildflower meadow. 16:9. 4K. 24fps. SOUND ON. ONE CONTINUOUS SHOT. NO CUTS. NO EXTREME MACRO. NO EXTREME CLOSE-UP. NO SLOW MOTION. NO CGI OR RENDERED APPEARANCE. The image should look like REAL wildlife footage captured outdoors with a professional cinema camera. The bee is important, but the ENVIRONMENT is equally important. Do not fill the frame with the bee. Keep enough distance from the bee that we clearly see the surrounding meadow, flowers, vegetation, sunlight and atmospheric depth. The result should feel natural, imperfect and observational rather than like a commercial or computer-generated insect demonstration. ──────────────────────── ENVIRONMENT ──────────────────────── A lush natural wildflower meadow during a warm spring morning. The background is rich and detailed: tall green grass, small yellow wildflowers, white daisies, lavender flowers, tiny weeds, different plant species, uneven vegetation, distant trees, soft rolling countryside, patches of blue sky, thin atmospheric clouds. Morning sunlight passes through the vegetation. Some flowers are brightly illuminated while others remain naturally shaded. A gentle breeze creates continuous subtle movement throughout the environment. Grass sways independently. Flower stems bend slightly. Leaves tremble. Small plants move at different speeds depending on their size and distance from camera. Occasional tiny insects appear naturally far in the background. Nothing should look perfectly arranged. This is a REAL meadow, not a studio flower arrangement. ──────────────────────── BEE APPEARANCE ──────────────────────── One normal-sized REAL honeybee. Accurate natural proportions. Natural brown, amber and muted golden coloration. Soft fine body hairs. Dark compound eyes. Transparent wings. Realistic segmented body. Six anatomically correct legs. The bee should NOT look perfectly clean. Its body has subtle natural irregularities. A few tiny pollen particles cling to its legs and body. IMPORTANT: Avoid shiny plastic surfaces. Avoid glossy artificial materials. Avoid perfectly smooth geometry. Avoid excessive sharpness. Avoid metallic gold coloration. Avoid exaggerated black-and-yellow stripes. Avoid oversized eyes. Avoid cartoon proportions. Avoid CGI-style symmetry. The bee should look like an ordinary living honeybee photographed in nature. ──────────────────────── 0:00–0:05 β€” ENTERING THE MEADOW ──────────────────────── Begin with a beautiful medium-wide nature-documentary composition. Camera positioned approximately at flower height. Foreground: several softly out-of-focus blades of grass and wildflowers. Midground: the honeybee flies naturally between flowers. Background: a large meadow extending into the distance with trees and warm morning sky. The bee occupies only approximately 10–15% of the frame. DO NOT zoom into it. Camera gently tracks sideways while keeping the bee within the composition. The bee does NOT fly in a perfectly straight line. Its flight contains tiny natural corrections: slight rises, small drops, subtle left-right movements, brief hovering. Its wings move rapidly with natural motion blur. The wings should NOT appear frozen or individually animated. Sunlight occasionally catches the translucent wings. ──────────────────────── 0:05–0:10 β€” THROUGH THE FLOWERS ──────────────────────── Continue the same uninterrupted shot. Camera follows from several feet away. The bee travels deeper through the meadow. A large lavender flower passes close to camera in the foreground and becomes naturally blurred. The bee briefly disappears behind the flower for a fraction of a second. It emerges on the other side as the SAME bee. Focus gently adjusts. Do NOT snap focus unnaturally. Allow occasional slight softness. The bee approaches a cluster of small yellow flowers. It briefly hovers. Its legs hang naturally underneath its body. The environment remains alive around it. Grass moving. Flowers swaying. Leaves catching sunlight. A tiny insect passes far in the background. Natural atmospheric depth. Do NOT blur the entire background into abstract bokeh. We should still understand and SEE the meadow. ──────────────────────── 0:10–0:15 β€” NATURAL LANDING ──────────────────────── The bee chooses one yellow wildflower. It circles it slightly. Then gently descends. Camera remains at the SAME observational distance. NO dramatic push-in. NO zoom. The bee extends its legs. Touches the flower. The flower head bends very slightly under the contact. The bee settles naturally onto the center. Its wings stop beating and fold against its body. The flower continues moving gently in the breeze. Camera does NOT move closer. Instead, hold the wider composition. Bee resting on flower in the midground. Multiple layers of wildflowers around it. Tall grass moving. Warm sunlight. Distant trees. Natural sky. A second tiny bee passes very far in the background, barely noticeable. Hold this peaceful natural composition. CUT EXACTLY AT 15.00 SECONDS. ──────────────────────── CAMERA / IMAGE QUALITY ──────────────────────── 4K documentary cinematography. Professional full-frame cinema camera. Natural 70–100mm telephoto/mild macro lens characteristics. Camera stays several feet from the bee. NO extreme macro lens. NO microscopic perspective. NO extreme magnification. NO artificial zoom. NO dramatic push-in. Use moderately shallow depth of field, NOT extremely shallow depth of field. The bee should be readable and detailed while the surrounding flowers and environment remain recognizable. Natural optical softness. Realistic fine detail. Natural motion blur. Subtle sensor grain. Realistic highlight roll-off. Slight atmospheric haze. Realistic dynamic range. Do NOT oversharpen. Do NOT make every microscopic detail unnaturally crisp. Real wildlife footage often contains slight focus imperfections β€” preserve that feeling. ──────────────────────── LIGHTING ──────────────────────── ONLY natural outdoor lighting. Warm morning sunlight. Soft blue skylight. Natural shadows from vegetation. Occasional sunlight flickering across the bee as it passes between plants. No studio rim light. No artificial glow. No perfect product lighting. No HDR appearance. No excessive contrast. No orange-and-teal grade. Natural documentary color science. Greens should contain many subtle variations rather than one artificial saturated green. ──────────────────────── SOUND ──────────────────────── Natural meadow ambience. Soft wind moving through grass. Leaves rustling. Distant birds. Very subtle insects. Occasional bee buzzing only when physically appropriate. The bee should NOT produce an exaggerated loud cartoon buzzing sound. When it flies farther from camera, buzzing becomes quieter. When it briefly passes closer, buzzing becomes slightly more noticeable. When it lands, buzzing fades naturally. NO MUSIC. NO NARRATION. NO DIALOGUE. ──────────────────────── REALISM PRIORITY ──────────────────────── The most important requirement is: THIS MUST LOOK PHOTOGRAPHED, NOT GENERATED. Favor natural imperfection over artificial perfection. Real insect anatomy. Real plant variation. Real sunlight. Real atmospheric depth. Real camera behavior. Real motion blur. Real focus behavior. Real environmental movement. The bee should belong naturally inside the environment rather than appearing composited on top of it. Its lighting must exactly match the surrounding scene. Its shadows must match the sun direction. Its depth of field must match its physical distance from camera. ──────────────────────── AVOID ──────────────────────── NO plastic bee. NO glossy bee. NO metallic bee. NO CGI bee. NO 3D-rendered appearance. NO game-engine look. NO artificial symmetry. NO perfectly clean insect. NO cartoon yellow-and-black coloring. NO giant bee. NO anthropomorphic features. NO fake eyes. NO extreme macro. NO extreme close-up. NO camera attached to bee. NO microscopic eye detail. NO giant compound-eye shot. NO dramatic zoom. NO artificial focus pull. NO excessive bokeh. NO studio lighting. NO commercial product-shot lighting. NO HDR. NO excessive saturation. NO excessive sharpening. NO fake lens flare. NO frozen environment. NO identical flowers. NO duplicated vegetation. NO artificial flower arrangement. FINAL VISUAL TARGET: A beautiful piece of REAL 4K wildlife-documentary footage. A normal honeybee naturally flying through a huge living meadow. The viewer can clearly see the bee, but also feels the scale and beauty of the entire environment around it. The bee should look like something a wildlife cinematographer actually captured with a long lens β€” slightly imperfect, naturally lit, organically moving and completely integrated into the scene. NOT A CLOSE-UP OF A BEE. A REAL BEE WITHIN A REAL ENVIRONMENT. EXACTLY 15 SECONDS.
1
150
Blender just became 10x more powerful with AI. 🚨 Blender + GPT Astra = a completely different way of creating visuals. Β  Instead of choosing between traditional 3D and generative AI, you can combine both. Β  β†’ Build your scene in 3D Β β†’ Control the camera, depth, and composition Β β†’ Use AI to transform the result Β β†’ Create visuals that would normally take hours or days Β  This is where creative workflows are heading. Β  Comment β€œLAB” and I’ll show you how to do this.
168
39
528
28,727
AI slop is everywhere right now. That’s exactly why inside AI Creator Lab, we don’t just teach you how to generate, we teach you how to actually direct AI properly. Better prompts. Better visuals. Better storytelling. Better taste. The tools are easy to access. Knowing how to use them well is the skill. Join us here: skool.com/aicreatorlab/about
6
424
I struggled with AI consistency so you don’t have to. 🚨 This used to be my biggest frustration with AI filmmaking. You create the perfect character. Then you generate the next scene... And suddenly: β†’ Their face is different β†’ Their clothes have changed β†’ The environment doesn't match β†’ Your "same character" looks like a stranger Doesn't matter how good each individual shot is. If nothing stays consistent, it doesn't feel like a film. So I ran that same scene through three different reference setups before I found what actually fixedΒ  it :Β locking the character to a single reference image and tagging it directly in every generation, instead of just re-prompting and hoping. Pressing "generate" is easy. Building a sequence where the character, product, environment and visual language stay consistent across every shot, that's where the real work is. That's exactly why I built AI Creator Lab. Inside, I teach the actual workflows I use to build consistent characters, maintain identity across scenes, and turn individual generations into finished sequences instead of random clips. I already burned the weeks testing this so you don't have to. Join us inside AI Creator Lab β†’ skool.com/aicreatorlab/about Follow @Sebastien Jefferies 🧩 for more tech & AI insights
1
7
292
Sebastien Jefferies retweeted
AI images are getting dangerously hard to spot. 🚨 GPT Image 2.5 is here and this is getting ridiculous. Think about how fast we got here And the craziest part? This is only 3 years of progress. I tested the SAME portrait prompt: No fancy prompt engineering. Just the exact same prompt. If you want to start actually directing with AI, join us inside AI Creator Lab hereβ†’ skool.com/aicreatorlab/about
1
2
252
Same prompt on @grok πŸ‘‡
Seedance 2.5 on @higgsfield_ai Prompt: Generate a 30-second, 16:9, 24fps high-octane Eastern martial arts short film. High-budget cinematic live-action with AAA CG visual quality. No subtitles, no UI, no logos, no text, no voiceover, no background musicβ€”retain only realistic sound effects: roaring wind, sharp sandals skidding, silk whipping through air, katana blades slicing air, thunderous metal collisions, bamboo stalks snapping, heavy body impacts, whip pans, and fierce breathing. 【Visual Style & Characters】 Protagonist: A solitary female ronin in dark indigo-black traveling robes with crimson lining, hair tied back in a high messy ponytail, wearing silver forearm guards and carrying a dark steel katana. Cold, lethal, unyielding aura. Antagonists: Two distinct male dual-wielding shadow assassins in ash-gray and bone-white ninja cloaks with bamboo face masks. Gray Assassin A wields a heavy naginata (polearm) for long-range overhead sweeps and thrusts. Gray Assassin B wields twin kama (scythes) for rapid, ultra-close acrobatic slashes. Must remain two distinct individuals throughout. 【Combat Dynamics】 A relentless 2 vs 1 high-speed encirclement battle. Continuous hyper-speed ninja agility and swordplay from the very first frameβ€”no stance-holding, no eye-contact pauses, no dramatic posing. Every 0.2–0.6 seconds features a dynamic transition in attack angle, elevation, or direction. Every jump, flip, slide, and rotation is seamlessly integrated directly into an attack, block, or parry. 【Camera Work】 Feral, intimate, hyper-dynamic handheld camera work dragged through the fight. Incorporates low worm's-eye views, dynamic side tracking, short rotating orbits, fast dolly-ins, violent recoil jerks on heavy weapon hits, Dutch angles (45Β°/85Β°), and foreground occlusions (snapped bamboo, flying leaves). Camera occasionally lags behind the action slightly before snapping back, conveying extreme momentum. 【Environment】 A misty, dense bamboo forest at twilight with falling autumn leaves and scattered mossy stone lanterns. The environment reacts heavily to the fight: bamboo stalks sheared in half, stone lanterns shattered by missed strikes, fallen leaves swirling violently in weapon draft currents, and dust trails kick-started by skidding feet. 【Action Timeline】 0–6s: Immediate dive into battle. Naginata lunges from foreground with a 0.1-second near-miss speed ramp skimming protagonist's cheek, while Scythe Assassin attacks from the rear flank. Protagonist low-slides beneath the naginata shaft, performs a one-handed ground spin to deflect the scythes, and parries the returning naginata thrust. 6–12s: Continuous high-low pressure. Protagonist vault-steps off Naginata's pole shaft into a mid-air twist slash, forcing Scythe Assassin back, lands into a cartwheel dodge, and side-flips through an attack gap with a low ankle-sweep cut. 12–18s: Midpoint chaos. Surrounded in a tight pinch. Camera fast dolly-in to an orbital whip-around as protagonist drops flat under a synchronized cross-strike, explosive body spin disarming one scythe, and back-slashing the polearm wielder. 18–24s: Maximum speed eruption. Multi-directional redirection. Head-tilt dodge β†’ hilt-strike to pole shaft β†’ 360Β° backflip counter β†’ bamboo-stalk wall-rebound β†’ aerial down-slice. 24–30s: Climax execution. Protagonist slices through the center gap. Rapid double-slash mauls Scythe Assassin, blasting him into a bamboo cluster that splinters on impact. Protagonist instantly pivots, executing a full-body rotating heavy cleave against Naginata Assassin across swirling leaves, closing for a final decisive strike. Final 0.5s: Protagonist stands alone, sword lowered, leaves settling around her, fast dolly-in to her cold eyes, hard cut to black. 【Negative Prompt / Exclusions】 Stance-holding, slow-motion pauses, turn-based fighting, idling, prolonged staring, floating energy beams, magic glowing auras, UI, text, logos, subtitles, character merging, cloned extra limbs, studio backdrops.
Made with AI
1
2
269
Replying to @grok
@grok is getting seriously good at animation. 😱 The motion, consistency, and overall cinematic feel are improving fast. Here’s the exact prompt I used to create thisπŸ‘‡
2
2
3
646
Create an EXACTLY 30-second cozy hand-drawn animated sequence inspired by the warmth, domestic observation, detailed food animation, and gentle countryside atmosphere associated with classic Studio Ghibli films. 16:9. SOUND ON. NO DIALOGUE. NO NARRATION. NO SUBTITLES. NO TEXT. NO LYRICS. The sequence follows ONE elderly grandmother preparing and enjoying a very simple breakfast in her old countryside kitchen at sunrise. Nothing dramatic happens. The entire emotional experience comes from: FOOD, SOUND, LIGHT, SMALL MOVEMENTS, OLD OBJECTS, STEAM, WEATHER, AND THE GRANDMOTHER'S QUIET ROUTINE. The sequence should feel like a tiny memory from childhood. Warm. Nostalgic. Peaceful. Handmade. Intimate. Slightly melancholic. Most importantly: THE SOUND DESIGN IS HALF OF THE EXPERIENCE. Every tiny sound should be beautifully isolated and satisfying: CRUNCH of bread crust, knife scraping butter, wood creaking, gas flame igniting, coffee bubbling, metal lid trembling, ceramic clinking, coffee pouring, spoon stirring, birds outside, curtains moving, old clock ticking. No exaggerated cartoon sound effects. Everything should feel intimate and real. ──────────────────────── GRANDMOTHER β€” CONTINUITY LOCK ──────────────────────── ONE elderly grandmother approximately 75–80 years old. Small, slightly rounded build. Soft wrinkled face. Warm rosy cheeks. Kind dark-brown eyes. Silver-gray hair gathered into a loose low bun. Wispy strands around her temples and forehead. Visible age in her hands: wrinkles, slightly enlarged knuckles, tiny age spots, short natural nails. She wears: faded cream cotton blouse, dusty-blue knitted cardigan, long warm-brown skirt, old floral apron, thick wool socks, worn house slippers. Her clothing is slightly faded from years of washing. She moves slowly but confidently. These are movements she has performed thousands of times. She never looks toward the camera. She never performs for the audience. She simply continues her morning. ──────────────────────── THE KITCHEN ──────────────────────── A tiny countryside kitchen inside an old lived-in house. Early morning. Wooden cabinets with worn edges. Cream plaster walls. Small handmade tiles behind the stove. Copper pans hanging from hooks. Old metal kettle. Vintage gas stove. Small metal stovetop coffee pot. Glass jars containing: flour, sugar, dried herbs, coffee. Wooden cutting boards. Mismatched ceramic plates. Old kitchen towels. Small vase containing three wildflowers. A basket with several fresh eggs. Rustic loaf of bread wrapped in a faded cloth. Tiny ceramic butter dish. Blue-and-white coffee cup. Old wooden table covered with scratches from decades of use. Nothing matches perfectly. Nothing is decorative simply for aesthetics. Everything exists because someone uses it. A small open wooden window overlooks: green hills, vegetable garden, wildflowers, trees moving gently in morning wind. Thin cream curtains move in the breeze. Golden sunlight enters through the window. Dust particles drift through the beam. Occasional steam passes through the sunlight. The room should almost feel like it has a smell: fresh bread, coffee, old timber, cool morning air. ──────────────────────── 0:00–0:05 β€” BREAD ──────────────────────── OPEN directly on an intimate medium shot. No establishing shot. Grandmother is already sitting at the wooden kitchen table beside the open window. Morning sunlight illuminates the side of her face. She holds a thick piece of rustic bread in both hands. Outside: quiet birds. Soft leaves. Very distant countryside ambience. She raises the bread. Takes one slow bite. CRRRRUNCH. Make the crust sound extremely satisfying. Hold on the moment. Tiny crumbs break free. CUT to macro close-up. Several crumbs land on the scratched wooden table. tik... tik... One remains caught on her cardigan. She slowly chews. Very subtle: chew... chew... Her eyes drift toward the open window. A tiny breeze enters. Curtain moves. Her loose silver hair strands move slightly. Wooden chair: creeeak. She reaches toward a small butter dish. ──────────────────────── 0:05–0:09 β€” BUTTER ──────────────────────── EXTREME CLOSE-UP. Her wrinkled hand picks up an old butter knife. Metal touches ceramic: TING. She cuts into soft yellow butter. Very subtle: scrrrape... The knife drags slowly across the warm bread. SCRAAAAAPE. Focus on the texture. Crispy uneven crust. Soft airy interior. Butter catching on tiny holes. The warmth of the bread begins melting it immediately. A small glossy patch forms. She carefully reaches the knife toward the edge. One tiny bit of butter almost falls. She catches it with the knife. No comedy reaction. Just an everyday practiced movement. Knife placed gently onto plate. clink. She takes another smaller bite. CRUNCH. Thenβ€” from behind her: very faint... blup. She pauses chewing. Another: BLUP. Her eyes move toward the stove. The coffee is ready. ──────────────────────── 0:09–0:14 β€” COFFEE ON THE STOVE ──────────────────────── CUT to extreme close-up. Small old metal coffee pot over a blue gas flame. FSSSSSSSSHHHH... The metal surface has scratches and years of discoloration. Tiny reflections of the blue flame move across it. Inside: coffee begins bubbling. blup... ... blup... ... BLUP-BLUP... The metal lid starts trembling. tik... ... tik-tik... Steam escapes around one edge. psssss... Beautiful hand-drawn steam curls upward. It passes through golden sunlight. For a moment, the steam becomes luminous. Camera follows the steam slightly upward. Grandmother enters softly out of focus in background. She walks toward the stove. Her slippers: shff... shff... shff... Old floorboard: CREAK. Her wrinkled hand reaches into frame. Grips the wooden handle. Her thumb naturally adjusts around it. OTHER HAND: turns the gas knob. CLICK. FSSSSSSβ€” The flame disappears. Sudden quiet. Hold the silence for a fraction of a second. Only birds remain. This moment should feel incredibly satisfying. ──────────────────────── 0:14–0:19 β€” THE POUR ──────────────────────── CUT to table-level macro shot. Her handmade ceramic cup waits on the table. Cream-colored ceramic. Tiny blue painted flowers. Uneven handmade rim. Small chip near the handle. She places the coffee pot beside it. TOK. She sits. Chair: creeeak. Grandmother lifts the coffee pot. Very carefully tilts it. A dark stream of coffee begins pouring. SHHHHHHHHHHH... Make this one of the strongest sounds in the sequence. Extreme close-up. Coffee stream hits the bottom of the ceramic cup. Initially: sharp hollow liquid sound. As the cup fills, the resonance gradually becomes deeper and softer. Coffee swirls. Tiny bubbles collect around the edge. Steam rises. Dark brown liquid catches golden reflections from the window. The grandmother slightly changes the angle of the pot. The stream becomes thinner. shhhhh... shhh... ... stop. One drop remains suspended from the spout. Hold for a tiny moment. PLOP. Perfect circular ripple. Let the ripple expand across the entire coffee surface. ──────────────────────── 0:19–0:23 β€” SPOON ──────────────────────── Her fingers reach toward a tiny silver spoon. She lifts it. TING. A small sugar jar opens. Wooden lid: tok. One spoonful of sugar. Tiny crystals fall. sssshh... into the coffee. The spoon enters the cup. CLINK. Grandmother slowly stirs. cling... ... cling... ... cling... Macro overhead shot. Coffee rotates into a beautiful dark spiral. Tiny bubbles orbit the cup. Steam curls around her fingers. The spoon gradually slows. One final: ...clink. She removes it. A tiny coffee droplet falls from spoon back into cup. plip. She places spoon on saucer. ting. ──────────────────────── 0:23–0:27 β€” THE KITCHEN BREATHES ──────────────────────── Now finally move to a WIDER composition. Grandmother sits at the wooden table. Coffee steaming. Half-eaten bread beside it. Butter dish. Wildflowers. Morning sunlight. Open window. The room becomes almost completely still. Allow environmental animation to carry the scene. Curtain gently billows inward. A hanging bundle of dried herbs moves slightly. Steam curls above coffee. Dust floats through sunlight. Outside, leaves move. A tiny sparrow lands on the windowsill. tap-tap. Grandmother notices. She stops moving. The bird tilts its head. Grandmother tilts hers slightly too. No exaggerated smile. Her expression simply softens. Another bird chirps somewhere outside. The sparrow answers. Grandmother's mouth forms the smallest warm smile. ──────────────────────── 0:27–0:30 β€” FIRST SIP ──────────────────────── Close-up on her hands. Both wrinkled hands wrap around the warm ceramic cup. Tiny ceramic movement against table: rrrrk. She pauses. Feels the warmth through her fingers. Steam passes across her face. She gently blows across the coffee. fffffff... Steam bends away from her breath. She takes ONE small sip. Very quiet: sip. Places cup down. tok. Hold on her face. Eyes looking through the open window. Morning sunlight across her wrinkles. Peaceful. Satisfied. Nothing else needed. Thenβ€” the old kitchen clock somewhere behind her: TICK. ... TICK. A bird sings outside. Steam continues rising from the cup. CUT TO BLACK EXACTLY AT 30.00 SECONDS. ──────────────────────── CRITICAL SOUND DESIGN ──────────────────────── SOUND DESIGN IS THE MAIN CHARACTER. NO SPEECH WHATSOEVER. NO audible human conversations. NO radio. NO television. NO narration. NO lyrics. Prefer NO MUSIC AT ALL. Allow silence and domestic sounds to create the rhythm. Each sound should occupy its own acoustic space. VERY CLOSE SOUNDS: bread crust cracking, crumbs landing, knife scraping bread, butter knife touching ceramic, coffee bubbling, metal lid trembling, gas flame, pot touching table, coffee stream, sugar falling, spoon touching cup, coffee stirring, cup touching table, grandmother gently blowing, tiny sip. ROOM SOUNDS: wooden chair creaking, floorboards, fabric movement, cardigan sleeves, slippers, old clock. OUTSIDE SOUNDS: sparrows, leaves, gentle breeze, distant insects, very distant countryside ambience. NO loud cinematic impacts. NO whooshes. NO trailer bass. NO exaggerated anime sound effects. The audio should feel almost ASMR-like but completely natural. ──────────────────────── ANIMATION QUALITY ──────────────────────── Traditional hand-drawn Japanese feature-film animation with the warm domestic craftsmanship associated with classic Studio Ghibli filmmaking. Hand-painted watercolor and gouache backgrounds. Organic pencil construction. Soft imperfect ink outlines. Traditional cel-animation feeling. Beautifully observed everyday human motion. Food receives unusually detailed animation. Steam receives unusually detailed animation. Liquid receives unusually detailed animation. Hands receive unusually detailed animation. Do NOT make every element move constantly. Use STILLNESS. A curtain moving while everything else remains still is enough. Steam rising from an otherwise motionless frame is enough. Grandmother blinking once can become meaningful. Use animation on twos where appropriate. Occasional held drawings. Then use smoother animation for: coffee pouring, steam, cloth, hair, hand movements, birds, liquid. Her elderly body must have believable weight. When standing: hands briefly assist, torso leans forward, legs take weight, cardigan shifts afterward. When sitting: chair responds to weight. When reaching: shoulder initiates movement, elbow follows, wrist rotates, fingers articulate last. Avoid modern perfectly interpolated movement. ──────────────────────── FOOD & LIQUID ANIMATION ──────────────────────── The food should look irresistibly tactile. BREAD: thick rustic crust, uneven handmade shape, crisp exterior, soft porous interior, crumbs, slight compression when held. BUTTER: soft, slightly glossy, slowly melting into warm bread. COFFEE: deep warm brown, subtle reflected highlights, realistic viscosity, tiny bubbles, surface tension, ripples, steam. When coffee pours: animate the stream continuously. Do NOT make it look like water. It should feel slightly heavier and darker. When coffee fills the cup, SOUND must physically change as the liquid level rises. ──────────────────────── LIGHT ──────────────────────── Sunrise. The kitchen begins with soft warm morning illumination. Golden sunlight enters diagonally through the window. Use: warm honey highlights, cream, faded yellow, dusty blue, muted sage, terracotta, natural brown, soft green. Shadows are gentle and slightly cool. The grandmother's skin remains warm. When steam crosses sunlight, it glows almost white-gold. Dust particles should only become visible inside direct sunbeams. No excessive bloom. No modern HDR look. No aggressive saturation. ──────────────────────── CAMERA LANGUAGE ──────────────────────── Camera behaves like a quiet family member sitting nearby. Never flashy. Never invasive. Use approximately: 40–50mm character compositions. Macro inserts for: bread, wrinkled fingers, butter, coffee bubbles, gas flame, pouring stream, coffee ripple, sugar, spoon, steam. Movement should be extremely restrained. Small pans. Very slow pushes. Occasional completely static compositions. NO handheld shake. NO FPV. NO dramatic orbit. NO drone shot. NO rapid zoom. NO modern social-media editing. Allow shots to breathe. ──────────────────────── NEGATIVE CONSTRAINTS ──────────────────────── NO dialogue. NO narration. NO subtitles. NO text. NO lyrics. NO modern kitchen. NO espresso machine. NO smartphone. NO television. NO laptop. NO plastic modern appliances. NO photorealism. NO live action. NO 3D CGI. NO glossy modern anime. NO ultra-clean digital outlines. NO vector artwork. NO plastic-looking characters. NO chibi. NO exaggerated facial expressions. NO magical creatures. NO fantasy spectacle. NO dramatic plot. NO crying. NO slapstick comedy. NO excessive camera movement. NO hyperactive editing. NO loud soundtrack. Do NOT rush the sequence. Do NOT turn the grandmother into a caricature. Do NOT make every movement overly cute. Do NOT make the environment unrealistically pristine. FINAL TARGET: A 30-second animated sensory memory of an elderly grandmother's morning. She quietly eats warm bread. CRUNCH. Spreads melting butter across it. SCRAAAAAPE. Coffee begins bubbling behind her. BLUP... BLUP... She walks to the old stove. CLICK. The flame disappears. Silence. She pours coffee. SHHHHHHH... One final drop. PLOP. Sugar falls. Spoon touches ceramic. CLINK. She stirs slowly. cling... cling... cling... A sparrow lands at the window. She notices it. Her expression softens. She wraps both hands around the warm cup. Blows gently across the steam. Takes one small sip. The old clock ticks. Birds sing outside. Nothing extraordinary happens. And somehow it should feel like the most comforting 30 seconds imaginable. EXACTLY 30 SECONDS.
672
Sebastien Jefferies retweeted
Replying to @grok
@grok is getting seriously good at animation. 😱 The motion, consistency, and overall cinematic feel are improving fast. Here’s the exact prompt I used to create thisπŸ‘‡
2
2
3
646
AI just made traditional film production look ridiculously slow. 🚨 I made this video using GPT Image 2.5, and this is probably the part that impressed me most: One photo of a face β†’ keep that exact character across 20+ images. Then change the outfit, location, lighting or composition while keeping everything else consistent. You can basically build entire campaigns before lunch. Comment β€œLAB” if you want to learn how to use these tools to make art.
2
4
190