Made by Creators. Found on VEYR.
Entertainment executives at Variety’s September 17 summit argued for task-specific AI rules that protect rights and preserve accountable human decisions.
Read StoryThe live pulse of AI entertainment: creator posts, prototypes, workflows and fast-moving breakthroughs.
See all Signal
AKOOL has opened an AI Film Challenge spanning imaginary movie trailers, AI short films and an 'Impossible' category, with $500,000 in advertised creator rewards and a November 20 deadline.
Future Vision XPRIZE selected five finalists from more than 2,500 hopeful sci-fi film trailer and treatment submissions, with winners due September 25 at Moonshots LIVE.
Qwen-Image-2.1 combines generation and editing in one open model, adds native RGBA transparency, supports up to 10 reference images, and can make targeted local edits.
A 90-second AI short called The Last Promise uses a grounded samurai-style duel and restrained effects as its creative test, with the creator emphasizing physical choreography and impact.
Runway's API now includes frame-rate enhancement for clips up to five minutes, with targets from cinema frame rates through 120fps.
Higgsfield’s Cinema Studio 4.0 adds director-style controls for genre, camera, pacing, references, color and lighting, turning AI video generation into a more structured filmmaking workflow.
EVENT WATCH upcoming
Runway's one-day San Francisco summit will bring together AI researchers, builders and executives, with world models, video and new gaming experiences on the agenda.
Event hubDeeper production methods, multi-tool processes and creator case studies.
See all workflows
Autodesk used AU26 to demonstrate an AI-assisted workflow that keeps characters, performances, cameras, environments, and animation editable.
A two-person first-time game team used AI to move from idea to implementation, then found that integration and system coherence became the harder problem as the project grew.
Pocket FM launched Sherpa, an AI fiction system that coordinates specialist agents and long-term narrative memory for serialized stories.
GMI Cloud’s Summer Signal event passed one shared artifact through six multimodal transformations, framing image, voice, video, world models and agents as a connected production system rather than separate demos.
Veteran Nintendo designer Takaya Imamura says iterating with AI on a 3D shooter prototype feels like his old back-and-forth with programmers, highlighting how agentic tools can let non-programmer designers reach a playable test faster.
LIVE 404 led a field of 45 teams at Tokyo AI’s rapid builder challenge, where working games and creator tools put AI inside play and production.
View original source
A creator with no coding background says about 200 Astra-led iterations produced DragonWild, a playable browser prototype built around an emergent dragon-world simulation.
View original source
Robert Gaudette's Runway Grand Prix short has now won the 2026 Reply AI Film Festival in Venice, selected from more than 3,000 submissions across 76 countries.
View original sourceImportant new models and updates for creators.
OpenAI will discontinue the Sora Videos API and Sora 2 models on September 24, creating an immediate migration deadline for automated creator pipelines that still depend on them.
Former Google DeepMind Genie researchers formed Emulate to pursue world models for simulated environments, with major fundraising discussions reportedly underway.
Google launched agentic video understanding for Gemini, dynamically scanning video segments while cutting token use and cost.
CapCut introduced Dreamina Seedance 2.5 and Dola Seedream 5.0 Pro with longer clips and more precise editing controls.
The complete VEYR news archive.
Runway's API now includes frame-rate enhancement for clips up to five minutes, with targets from cinema frame rates through 120fps.
Future Vision XPRIZE selected five finalists from more than 2,500 hopeful sci-fi film trailer and treatment submissions, with winners due September 25 at Moonshots LIVE.
Qwen-Image-2.1 combines generation and editing in one open model, adds native RGBA transparency, supports up to 10 reference images, and can make targeted local edits.
AKOOL has opened an AI Film Challenge spanning imaginary movie trailers, AI short films and an 'Impossible' category, with $500,000 in advertised creator rewards and a November 20 deadline.
Higgsfield’s Cinema Studio 4.0 adds director-style controls for genre, camera, pacing, references, color and lighting, turning AI video generation into a more structured filmmaking workflow.
A 90-second AI short called The Last Promise uses a grounded samurai-style duel and restrained effects as its creative test, with the creator emphasizing physical choreography and impact.
OpenAI will discontinue the Sora Videos API and Sora 2 models on September 24, creating an immediate migration deadline for automated creator pipelines that still depend on them.
A creator with no coding background says about 200 Astra-led iterations produced DragonWild, a playable browser prototype built around an emergent dragon-world simulation.
After a viral trailer drew criticism over AI art, Tooth for Tooth creator Alex Kunchevsky published a detailed account of where AI assisted the game's production pipeline.
Slapjam opens September 28 as a free 48-hour online jam for portrait-format browser games made with disclosed AI tools.
A two-person first-time game team used AI to move from idea to implementation, then found that integration and system coherence became the harder problem as the project grew.
Mojiworks’ playable Reddit beta lets players describe proposed buildings, see AI visualize them in a shared town, and vote on what becomes part of the persistent world.
Curious Refuge compares three AI methods for creating alternate camera angles and publishes the accompanying prompts and written breakdown.
Franklin AI says a Tripo Smart Mesh test replaced 2D placeholders with more than 35 textured 3D assets for a Mario Kart-inspired prototype, then integrated them through Claude Code.
Franklin AI tests TypeSafe AI’s Jev decision model with Doom, showing how typed, low-latency outputs could support real-time game behavior without free-form generation.
A live Reddit word puzzle uses hand-authored fusion recipes first and calls Gemini only for combinations the creator has not predefined.
LIVE 404 led a field of 45 teams at Tokyo AI’s rapid builder challenge, where working games and creator tools put AI inside play and production.
Bahman Ghobadi’s Eva Green-led psychological horror feature premiered at TIFF, and its generative-AI imagery has become a focus of critical debate.
Pocket FM launched Sherpa, an AI fiction system that coordinates specialist agents and long-term narrative memory for serialized stories.
TheGrill’s September 30 Los Angeles program will bring filmmakers and toolmakers behind Terrarium, ANCESTRA, and Young Washington onto one hybrid-AI panel.
Disney named former Character.AI chief Karandeep Anand its first companywide CTO, placing data and AI platforms inside a broad enterprise role.
Entertainment executives at Variety’s September 17 summit argued for task-specific AI rules that protect rights and preserve accountable human decisions.
Former Google DeepMind Genie researchers formed Emulate to pursue world models for simulated environments, with major fundraising discussions reportedly underway.
Autodesk used AU26 to demonstrate an AI-assisted workflow that keeps characters, performances, cameras, environments, and animation editable.
System 3 says confusion over AI-assisted James Pond Legacy art began with contractor wording that described a model step while calling final assets human-authored.
CESA says 85.8% of surveyed Japanese game developers use generative AI at work, alongside common rules for human review and output controls.
Theoretically Media showcases a cinematic AI film workflow using GPT-6, Codex, Runway, and DaVinci Resolve to remake a film in about a day and a half
A solo developer is testing how far GPT-6 Astra and Meshy v7 can carry a playable isometric sci-fi ARPG prototype inside Unreal Engine 5.8.
Tilly Norwood’s live press-tour experiment shows how a synthetic screen character can become an interactive promotional performer, along with the privacy and reliability costs.
Fotor launched an agentic video-production system that turns a brief into a fully assembled, editable multi-track project, including long-form narrative video and motion graphics.
After criticism of AI-generated material in its VISION 2026 II presentation, Level-5 CEO Akihiro Hino apologized for overusing AI in the showcase while saying scenarios, character designs, and foundational settings remain human-made and AI is being pursued mainly for production efficiency.
A Shanghai court ordered an AI voice service to stop selling packs modeled on 63 Genshin Impact characters and pay miHoYo 750,000 yuan after finding infringement and unfair competition.
GMI Cloud’s Summer Signal event passed one shared artifact through six multimodal transformations, framing image, voice, video, world models and agents as a connected production system rather than separate demos.
Veteran Nintendo designer Takaya Imamura says iterating with AI on a 3D shooter prototype feels like his old back-and-forth with programmers, highlighting how agentic tools can let non-programmer designers reach a playable test faster.
The inaugural African AI Film Festival requires creators to disclose the AI tools and models used in every entry while judging story, craft, originality and cultural resonance rather than penalizing disclosure itself.
Creator workflows combining Tripo’s 3D models with GPT-6 Astra, Blender and Codex are moving AI 3D beyond static assets into rigged, animated and interactive production.
Tamar AI launched Glanze at the Venice Film Festival as a distribution and discovery platform for AI-native series, films, and explorable worlds, opening with 40 preview titles.
Kerri Zane Enterprises says Higgsfield Cinema Studio cut concept and pre-production time 25–40% and early visualization costs 30–50% by letting the team test moving creative directions before committing production resources.
ElevenLabs released a full audiobook of The Odyssey narrated by the official AI voice of Sir Michael Caine.
At CapCut World Singapore, the company said more than 7 million creators use its AI features every day.
CapCut introduced Dreamina Seedance 2.5 and Dola Seedream 5.0 Pro with longer clips and more precise editing controls.
Stan Lee Universe partnered with ElevenLabs to bring an officially licensed AI voice and related creative templates to the platform.
The AI International Film Festival is now running monthly Hollywood screenings with blind judging and on-demand winners.
.webp)
Hasbro and ElevenLabs made officially licensed voices including Optimus Prime and Cobra Commander available for approved AI use.
Google says MrBeast is using Gemini and Google Health tools to brainstorm and execute ambitious video concepts, including extreme-environment challenges.
.webp)
ElevenLabs' Dubbing v2 is designed to preserve emotion, pacing and delivery across more than 90 languages.
Unity's AI tools entered open beta with an in-editor assistant, agent mode, MCP server, gateway and generators.
Google launched agentic video understanding for Gemini, dynamically scanning video segments while cutting token use and cost.
Adobe's new Generative Media workflow creates editable video and audio directly inside Premiere, while After Effects gets its AI Assistant in public beta.
AI Film 3's AI Entertainment Conference will run Sept. 17–19 with film, gaming, art, fashion and music programming.
Shift Up leadership said it is taking backlash over AI-generated promotional work seriously while maintaining that AI will influence game development.
Wonder Studios scheduled private screenings of KITA by Jidenna and Goldfish Mayhem by Junie Lau for its Sept. 10 LA premiere at The Lighthouse in Venice, with a filmmaker panel and industry reception.
Roblox announced Build, a mobile-first creation experience plus new agentic tools for professional creators.
Robert Gaudette's Runway Grand Prix short has now won the 2026 Reply AI Film Festival in Venice, selected from more than 3,000 submissions across 76 countries.
Roblox's Cube foundation model now generates functional 3D objects that can behave interactively inside games.
CapCut launched an evergreen partner program offering AI creators tools, credits, early access, event invitations and monetization opportunities.
Playco says GPT-6 Astra reduced manual fixes by 50% across three themed game prototypes built from one grey-box foundation.
Runway's one-day San Francisco summit will bring together AI researchers, builders and executives, with world models, video and new gaming experiences on the agenda.
Made something exceptional with AI? We want to see it.
Submit