AI News
Artificial intelligence and machine learning in games
Freya Holmér follows up her AI-doom panel with a longer conversation about how generative tools may erode meaning in art and game development. The discussion centers on whether AI changes …
arXiv cs.GR details DiffusionShadow, a diffusion-based cache for neural volume rendering shadows. The method packs many precomputed shadow INRs into one model conditioned on light direction, cutting storage while keeping …
arXiv cs.GR details SUCRe, a graph transfer learning method that tries to avoid negative transfer by only moving over reliable knowledge. The paper combines uncertainty-aware matching with contrastive sampling, aiming …
arXiv cs.GR details a new speech-driven 3D facial animation method that aims to make lip motion look more physically believable. The model separates visible articulation into spreading, opening, and protrusion, …
The Cherno shows how AI helped finish Hazel’s multiplayer server work, turning a long-running backend task into something shippable. For teams building online games, the practical takeaway is less about …
GameFromScratch reports that Meta has unveiled two upcoming "game engines" at Meta Connect 2026: Horizon Create and Horizon Studio. The pitch points toward AI-driven creation tools for Meta's Horizon ecosystem, …
GamesIndustry.biz has announced the full line-up for its HR Summit 2026, with talks on AI, neurodiversity, unions, and workplace culture. The one-day event lands in London on October 1 and …
GamesIndustry.biz reports that Meta is pushing AI-assisted creation and new VR hardware in the same week. Horizon Create and Horizon Studio can generate playable 2D and 3D games from text …
Catlike Coding Godot adds a variant smooth01() helper for value-noise sampling in Godot shaders, letting developers choose linear, smooth, or smoother blending. The practical win is control: use smoother interpolation …
SIGGRAPH has unveiled the keynote lineup for SIGGRAPH Asia 2026, putting Markus Gross, Ran Huang, and Andrea Vedaldi on stage. For graphics programmers and technical artists, that usually signals a …
arXiv cs.GR details SsgCaps, a public dataset for benchmarking sound scene generation. The set uses human-engineered scenes tied to structured prompts and only public-domain audio, making it easier for teams …
arXiv cs.GR details ARS-Avatar, a surfel-based method for animatable, relightable human avatars from multi-view images under unknown lighting. The system adds deformation priors, deferred shading for BRDF estimation, and learnable …
GamesIndustry.biz details the EU Commission’s draft KIDS Act, which could bring broad new rules to online games sold in EU territories. Developers may need stronger age assurance, parental controls, and …
arXiv cs.GR details AgriGen, a ROS-integrated scene generation framework on Isaac Sim for photorealistic agricultural robotics simulation. It procedurally builds row crops, orchards, and vineyards with physics, rendering, and domain …
arXiv cs.GR details PartLLM, a multimodal model that treats 3D part segmentation as intent-conditioned generation. For graphics and tools teams, the practical draw is one system that can handle text-guided, …
arXiv cs.GR details VideoX-Qwen, a data-centric video editing pipeline that pairs instruction data with a unified editor. The system builds 1.2 million directional edit records and reports strong results against …
GDC Festival of Gaming spotlights how Keeper’s bird companion flight stayed within budget by balancing pathfinding, locomotion, and animation. The session focuses on the memory tradeoffs behind fixed-wing movement, a …
NVIDIA Graphics says DLSS 5 is now available in NBA 2K27, bringing 3D-guided neural rendering and new art-direction controls to GeForce RTX 50 Series GPUs. The update aims to preserve …
arXiv cs.GR details an edge-centric brain transformer that models rs-fMRI connectivity as connections, not just regions. The framework uses edge time-series, a line graph, and a structure-aware transformer to improve …
arXiv cs.GR details MoSAT, a motion-synthesis model that uses spatial audio plus text to drive full-body animation. The team pairs it with STAM, a new dataset of motion sequences, directional …