Skip to content

Fields

Art & AIGC latest news

Generative AI in the art pipeline: concept art, textures, UI and marketing assets, and the copyright and jobs debates it raises.

118 picks73 in the last 30 days278 collected in all

Updated

Picks archive · Page 2

Nov 13Fri21–40
  1. Unreal Engine61

    Epic technical artist shares workflow for cleaning up Unreal Engine material graphs with AI and Unreal MCP

    Epic Games senior technical VFX artist Philippe Patient used Codex with Astra, connecting to Unreal Editor through the experimental Unreal MCP, to clean up a material graph he had spent half a day building: laid out left to right, grouped into labeled comment boxes, and values used in several places routed through named reroutes, with no parameters or calculations changed.

    Why it matters: An Epic technical artist shares prompts you can reuse as-is and a way of stating constraints that leaves the logic untouched, and the workflow carries over to other agent tasks inside an editor.

Nov 12Thu
  1. QbitAI59

    MiniMax open-sources image model H3-Image with native 4K and up to 16 reference images at 0.2 yuan per image

    MiniMax open-sourced H3-Image, an image generation model that supports native 4K output (16MP, 4096×4096 pixels) and accepts up to 16 reference images. Generating one 4K image on the web takes roughly 10-20 seconds in practice, the API price is 50% to 70% of mainstream competitors, and 16MP requests get another 50% off in the first week. In hands-on tests the author showed the model can infer front, side and top views from a product appearance image, generate mechanical exploded diagrams laid out according to assembly logic, and produce concept art for 3D game combat scenes that follows tactical instructions such as cover and suppressive fire.

    Why it matters: The author's hands-on tests cover CAD three-view breakdowns and 3D game scene concept art, which makes it possible to judge how usable the image model is in asset and concept design work.

  2. Chuapp72

    AI clones Feicui Gua Gua Gua demo a week after release as indie developer describes the difficulty of defending it

    Shimian Games released the Feicui Gua Gua Gua demo on a game jam platform on May 16; four or five days later, clone mini-games with nearly identical gameplay and visuals, converted to 3D by AI, appeared on Douyin and other platforms, and in July and August highly similar unreleased titles turned up on Steam as well, while the original game still has no confirmed launch date.

    Why it matters: Using the concrete experiences of developers such as Shimian Games, it lays out the real cost of AI cloning an indie game, along with the obstacles at the platform appeal and software copyright registration stages.

  3. Grok Bot Research52

    Ludo launches My Animations to generate reusable motions from a single sentence and mirror them onto any humanoid character

    Ludo has launched My Animations and mirrored animations, in which typing a single motion description generates a reusable animation saved to the account and playable on any character recognized as humanoid. Generation supports 1 to 10 second durations and 1 to 4 variants at 0.5 credits per animation, with advanced options such as looping, maintaining facing direction, start and end idle poses, and left and right hand poses.

    Why it matters: The piece lays out the full integration path for generating reusable motions from text along with the 0.5 credits unit price, so teams can judge whether it can replace their existing motion asset pipeline.

  4. Ars Technica69

    Big Walk fixed a bug with an LLM, and the indie game world is split over AI use

    Developer House House has publicly said it used an LLM in a limited way late in Big Walk's development to address a Unity engine performance bug involving infinitely long shadows, and that this code accounts for less than 0.1% of the game's code.

    Why it matters: The piece sets three positions side by side — jurors who declined to vote, the developer's own disclosure, and negative player reviews — showing where award juries actually draw the line on generative AI.

Nov 11Wed
  1. PocketGamer.biz42

    Layer launches Motion Designer Agent for chat-driven UA videos

    On November 2, Layer launched Motion Designer Agent, which lets game marketing teams describe what they need in a conversation and get back user acquisition (UA) videos ready to run. The agent writes the script, plans shots, pulls in assets, adds voiceover, music and sound effects, and handles animation, transitions and final compositing; Layer estimates about 15 to 30 minutes from creative concept to finished video.

    Why it matters: The article gives an estimate of roughly 15 to 30 minutes from idea to finished cut, plus how Creative Units are billed — useful benchmarks for judging this conversational UA production workflow.

  2. 80 Level47

    Blender 5.3 ships with native 3D Gaussian Splatting and DLSS denoising

    Blender 5.3 is out, with native 3D Gaussian Splatting support: splats can be imported from PLY, SPZ and glTF files, processed with Geometry Nodes, and rendered directly in EEVEE or Cycles.

    Why it matters: The article details Blender 5.3's native support for 3D Gaussian Splatting and DLSS denoising, giving art pipelines what they need to judge whether to upgrade.

  3. GameLook43

    Thai studio shows Bangkok Heat demo: eight people and eight days on the self-developed AI engine Oriverse produce 100 square kilometers of Bangkok

    Thai indie studio Earthshine Games posted a gameplay demo of Bangkok Heat on Reddit on November 3, saying its team of eight built 100 square kilometers of Bangkok streets in eight days with its self-developed AI engine Oriverse.

    Why it matters: Set against the solo development of Kingdom Rebirth, it shows which parts of the eight-day build were handed to AI and which still rested on years of accumulated work.

  4. Youxituoluo57

    Are game designers turning into "gacha pullers"?

    Drawing on the Hupu AI Workshop, a TapTap developer postmortem and interviews with several designers, the article asks whether AI lowering the barrier to game production will turn designers into "gacha pullers" who repeatedly generate and filter outputs, as the short-drama industry did. On November 4 and 5, the top fifteen spots on Hupu's trending list were all but taken over by small games from individual creators; meanwhile, a developer who used AI to build a 100-level Water Sort draws only about 100 organic impressions a day, two weeks after launch.

    Why it matters: From the Hupu AI Workshop to the rapid trial-and-error cycles of Turkish studios, the piece compares concrete shifts in designer roles and hiring standards as the barrier to entry falls, offering a lens on how AI is entering production pipelines.

  5. Gcores69

    19 IGF judges sign joint statement against generative AI games; Big Walk may be ruled ineligible over AI use

    Nineteen judges for the 2027 Independent Games Festival (IGF), including Sam Barlow and Austin Walker, issued a joint statement announcing they will refuse to vote for any game made with generative AI, and called on the IGF to draw up a strict policy that would ban all entries containing AI-generated content from competing.

    Why it matters: The stance of the 19 IGF judges who signed the joint statement against AI games, together with Big Walk's response acknowledging LLM assistance, is one sample of the conflict between award rules and disclosure obligations.

  6. fal44

    fal updates Meshy Rigging to turn humanoid 3D models into animatable characters

    fal has launched its Meshy Rigging update, which turns humanoid 3D models into characters that can be used directly for animation. The feature works with tools including Unreal Engine, Unity, Blender and Mixamo; finger rigging costs nothing extra, and multiple animations can be stacked on the same character.

    Why it matters: The update spells out which animations the rigged output supports and which engines it works with, so readers can judge whether their existing 3D assets can plug straight into an animation pipeline.

  7. 80 Level34

    Bloody Hell Hotel developer on vampire hotel sandbox design and procedural dungeons

    In an interview, Bloody Hell Hotel developer Wlad Marhulets laid out the design thinking behind the vampire hotel sandbox and confirmed the game will skip Early Access, launching directly as a 1.0 release on PC and consoles in 2027.

    Why it matters: The developer explains how tiered, controlled randomization keeps procedural dungeons from feeling repetitive, and discusses what it is like for a small studio in the age of generative AI.

  8. Unreal Engine61

    Epic turns Fab static assets modular inside UE 5.8 with MCP and Claude Code

    Epic Games solutions architect Franck Debaene used UE 5.8's MCP server (experimental) and toolset together with Claude Code to split a single-mesh static asset from Fab into reusable modules inside the editor, then rebuilt it along splines with a PCG graph into a desk with drawers that can be added or removed and a tray with planks that can be added.

    Why it matters: The entire workflow stays inside the editor, from splitting the single mesh to rebuilding it with a PCG graph, and the way the skill files are organized can be reused as is.

Nov 10Tue
  1. QbitAI57

    Fable 5.2's first hands-on tests: users build playable 3D games, 3D worlds and videos

    After Fable 5.2's release, early testers and researchers used it directly to generate playable 3D games, 3D worlds, videos and music; Wharton professor Ethan Mollick ran about 25,000 agents to generate more than 450 3D worlds.

    Why it matters: The piece rounds up hands-on cases, benchmarks and pricing data from several developers, so you can compare the new model's actual output and cost in game and 3D content generation.

  2. Roblox DevForum Announcements64

    Roblox opens beta for automatic translation of text in in-game images

    Roblox has launched a beta of automatic in-game image text translation, which translates text in images into the player's language while matching the original font, size and color as closely as possible; non-text elements stay unchanged. It currently supports French, German, Indonesian, Italian, Korean, Portuguese and Spanish. The feature is free and off by default, and must be enabled per language on the Localization page of the Creator Hub.

    Why it matters: The original lays out how to turn on automatic image translation, which languages are supported and the latency range, so developers can gauge which parts of their localization workflow need to change.

  3. Unreal Engine40

    Epic showcases Botanical Converter for region-accurate trees with Unreal MCP and PVE

    Epic Games senior technical product manager Oyvind Fiksdal showcased Botanical Converter, which turns a single prompt into a PVE preset after retrieving plant data through Unreal MCP, generating trees that match a region's vegetation.

    Why it matters: The article splits the pipeline from a single prompt to a PVE preset into four steps and shares advice on the pitfalls, making it a reference for building similar generation tools inside the engine.

Nov 9Mon
  1. Unreal Engine68

    Unreal Engine 5.9 MegaJam Preview released, the first UE version built primarily with agentic workflows

    The Unreal Engine 5.9 MegaJam Preview is out, and Epic says it is the first Unreal Engine release built primarily with agentic workflows. The preview opens up several experimental features, including an expanded toolset in the Unreal MCP extension, direct use of AI coding agents in the editor, AI-assisted content creation, and running projects in a browser.

    Why it matters: The UE 5.9 preview lists agentic workflows as a primary way of building and details experimental features such as AI coding agents inside the editor and running projects in a browser, which teams can use to judge the pace of adoption.

  2. Hyper3D Rodin55

    Hyper3D opens its Rigging API to all users

    Hyper3D by Deemos announced that its Rigging API is now available to all users, with documentation at https://docs.hyper3d.ai/en/api-specification/rigging. The company previously described the feature as automatically rigging humans, animals and fantasy creatures, down to individual fingers, with bones added or removed via prompt. It also offers 250+ motion presets and animation controls, so rigging, adjustment and animation all happen on one platform.

    Why it matters: The Rigging API is now open to all users, and the accompanying documentation lists request parameters and skeleton template options, so readers can gauge the cost of wiring it into an existing 3D pipeline.

Nov 6Fri
  1. GamesBeat40

    Nvidia announces DLSS 5 support for Hogwarts Legacy, War Thunder and Where Winds Meet

    Nvidia has announced that Hogwarts Legacy, War Thunder and Where Winds Meet will add DLSS 5 support over the course of November, with the Hogwarts Legacy update already released.

    Why it matters: The three studios' differing choices around DLSS 5's model controls show how much room developers have to tune between environmental detail and character consistency.

  2. Hyper3D Rodin39

    HYPER3D Rigging launches, turning a static model into a fully animated character in 15 minutes

    Hyper3D by Deemos has released HYPER3D Rigging, which automatically rigs a static model into a fully animated character, with an official workflow time of 15 minutes. The tool auto-rigs humans, animals and fantasy creatures down to each finger, offers 250+ motion presets and animation controls, and supports prompt-based editing in the rigging step — for example, typing "Add tail bones." adds tail bones. The post says game developer @HOYO_MAX used it to finish character animation.

    Why it matters: The post lists the creature types covered by automatic rigging and 250+ motion presets, which you can weigh against your own project's character count and rigging workflow.