Generative Ai News
The latest Generative Ai coverage curated for game developers.
arXiv cs.GR details a causal study showing audio-video generators can latch onto visual shortcuts instead of the event itself. The paper argues shared-latent designs don’t solve the problem, and that …
arXiv cs.GR details Mira-Scene, a 3D scene reconstruction method that aligns generated objects to pixels instead of relying on sparse pose guesses. For game teams building scene tools or AI-assisted …
arXiv cs.GR details WorldCrafter, a video world model that uses implicit 3D-aware memory to keep scenes consistent across viewpoints and long horizons. The system lets the requested camera angle shape …
GamesIndustry.biz says Tokyo Game Show 2026 felt more like a celebration of legacy IP than a preview of the next decade. The floor was packed with remakes, sequels, and franchise …
Freya Holmér is joining an AI doom panel discussion, and she says she’ll argue that generative AI is an anti-humanity technology. The panel is hosted and moderated by Whick-TV, with …
Game Developer reports that miHoYo won about $112,000 in a Shanghai court case after an AI voice service imitated Genshin Impact characters. The ruling could become a useful signal for …
arXiv cs.GR details a pipeline that turns figurative paintings into multiple plausible 3D interpretations instead of forcing one reconstruction. It samples camera-orbit videos, rebuilds them with 3D Gaussian Splatting, then …
arXiv cs.GR details a model that inserts 3D humans into partial graphic designs, aiming to make layouts feel more deliberate and visually coherent. The system predicts pose, framing, and placement …
arXiv cs.GR details PhysStream, a video-generation model that lets creators steer physics-driven motion mid-generation instead of locking in a full control schedule up front. For game teams, the interesting bit …
A new multi-agent LLM pipeline, ViFA-Council, uses GPT-4o, Gemini 3.1 Pro, and Claude Sonnet 4.6 to generate Vietnamese folk-art imagery and educational stories with tighter cultural control. The system adds …
Researchers have introduced Grow3D, a hierarchical octree-based method for generating 3D shapes with higher fidelity and faster inference. The approach predicts the next scale of an octree step by step, …
GamesIndustry.biz details Tim Willits saying Saber Interactive is experimenting with generative AI, but not using it in core production pipelines. He also says the studio has shifted development out of …
A new template-generation model, InterIL, jointly creates a background image and foreground layout from text instead of building them in sequence. For UI, marketing, and tools teams, the practical hook …
SceneHI brings high-resolution, 3D-consistent scene texturing with controllable illumination into a single generative pipeline. The system lifts priors from 2D diffusion models onto 3D objects, then bakes geometry-aware shadows directly …
GamesIndustry.biz reports that nearly 1,900 Blizzard workers have ratified their first union contract with Microsoft after more than two years of bargaining. The deal covers AI safeguards, crediting, remote work, …
Blizzard union workers have ratified a first contract covering about 1,900 employees, locking in guardrails around generative AI, layoffs, and remote or hybrid work. For game teams, it’s a notable …
GamesIndustry.biz reports that China’s games market topped $50.0 billion in 2025, reaching $51.8 billion after 5.4% year-over-year growth. Niko Partners expects another 4% gain in 2026, with mini-games, short-video discovery, …
Tim Ruswick argues that tool shaming is getting in the way of healthy game dev discourse, especially when developers attack each other over engines, assets, or workflow choices. The broader …
A new training-free pipeline turns plain text prompts into calligrams that preserve both shape and legibility. Telligram splits the problem into semantic occupancy and glyph reconstruction, using diffusion-guided skeleton optimization …
LightBridge aims to make relighting 3D Gaussian Splatting assets a one-pass process instead of a per-scene optimization job. The system uses a feed-forward generative pipeline plus a new relighting dataset …