Latest news with #Meshy
Forbes
06-05-2025
- Business
- Forbes
Move Over, Chatbots: The Next Tech Superstars Are 3D Foundation Models
Chatbots and 3D large models By author When ChatGPT burst onto the scene, it transformed how we interact with technology. But now, another AI revolution is quietly reshaping industries, promising not just smarter conversations, but immersive worlds—welcome to the era of 3D foundation models. These powerful AI systems generate intricate 3D models, animations, and virtual environments in mere moments, redefining how businesses operate from gaming and e-commerce to industrial design and beyond. This isn't just a creative tool; it's a strategic pivot with trillion-dollar potential. The 3D Content Crisis and AI's Answer The global thirst for 3D content is insatiable. The rise of virtual reality, augmented reality devices like Apple's Vision Pro, and digital twins—virtual replicas of physical objects used extensively in manufacturing and construction—have placed unprecedented demands on digital content creation. Traditional methods are too slow, expensive, and reliant on scarce expertise. Enter 3D foundation models: AI trained on massive databases of shapes, textures, and spatial relationships. These models rapidly convert sketches, photographs, or simple text prompts into fully-realized, production-ready 3D assets. "We're witnessing the democratization of spatial design," a AI engineer tells me. He spoke on condition of anonymity because he is not authorized to publicly comment on their product. "Smaller studios and global enterprises alike can now do in minutes what previously took weeks." The Rising Players in 3D AI The competition in 3D AI has intensified, dividing primarily between agile startups and tech giants aiming to dominate entire ecosystems. Here are some of the leading players in the field: Startups: Tripo: Favored by indie developers for its speed in generating editable 3D wireframes from single images. However, its lower texture detail opens opportunities for rivals. Rodin: Specializes in photorealistic avatars, granting users granular control over facial expressions and poses. Luxury brands and virtual influencers have already begun to embrace it. Meshy: Dubbed the "Shopify of 3D commerce," it optimizes models for real-time product visualization. But inconsistent texture quality reveals it's still maturing. Tech Giants: Microsoft Trellis: Aims at large-scale, enterprise-focused applications, especially mixed reality and industrial metaverse integrations. Strategic partnerships with Autodesk and Unity highlight its ambitious goals. Tencent Hunyuan 3D: Recently upgraded and open-sourced, it's become a versatile platform generating everything from game assets to intricate animations and consistent character sequences—ideal for industries reliant on storytelling. Video Generation: The Game-Changing Factor Among the most groundbreaking advances is FramePack, spearheaded by Lvmin Zhang, a pioneer behind the influential ControlNet project. FramePack combines the power of 3D modeling and AI video generation, making Hollywood-quality content accessible even on consumer-grade GPUs. This development doesn't merely shorten production timelines—it fundamentally disrupts traditional studio economics, affecting animation studios, advertising, and social media content creation. An LA-based visual effects producer currently trialing FramePack emphasizes, " 80% of lens dynamization needs can be handled with two RTX 4090 graphics cards at 1/5 the cost of traditional processes. This doesn't just reduce costs—it reshapes entire business models in entertainment and beyond." Navigating Geopolitical Realities Underpinning these technical advancements is a critical geopolitical layer. The escalating U.S.-China tech rivalry, exacerbated by stringent GPU export controls, has compelled Chinese tech giants such as Alibaba, ByteDance, and Tencent to push the boundaries of software optimization. Their focus on running complex AI models efficiently on consumer hardware demonstrates how innovation can circumvent traditional hardware barriers. "Efficiency is the new Moore's Law," notes a Silicon Valley venture capitalist closely observing these developments. "Those mastering lean, optimized 3D AI could dominate emerging markets and effectively sidestep sanctions." Real-World Impacts Across Industries The commercial and practical implications of 3D foundation models are staggering. Consider these scenarios: Retail: Virtual showrooms that previously took months to create can now be generated within hours. Healthcare: Detailed surgical simulations using AI-generated 3D organs significantly enhance training and preparation. Manufacturing: Instant prototyping through simple text prompts dramatically shortens product development cycles, reducing costs and accelerating innovation. Yet despite these promising developments, significant hurdles remain. Copyright and intellectual property disputes are already surfacing, as businesses grapple with the legal nuances of AI-generated content. Additionally, inconsistent quality in early outputs could lead to initial skepticism, reminiscent of the early days of ChatGPT. The Investment Imperative For investors, the 3D AI revolution represents an unprecedented opportunity. The smartest bets will be on companies that not only possess cutting-edge technology but also offer tailored, industry-specific applications—whether in gaming, architecture, virtual try-ons, or beyond. Businesses must heed this rapidly changing landscape urgently. Those unwilling or unable to adapt risk obsolescence as competitors embrace AI-driven spatial design. The next industrial revolution won't be written, it will be modeled. Indeed, while chatbots reshaped our communication, 3D foundation models promise to transform the very worlds—both physical and digital—that we inhabit.
Forbes
01-05-2025
- Entertainment
- Forbes
AI Game Development Enters Its Agentic Era With Upheaval's Dreamer Portal
Upheaval Games, a new startup formed by industry veterans from Blizzard, Tencent, Unity, and Obsidian, has launched early access to its flagship platform: the Dreamer Portal. It's a no-code, AI-driven tool that allows users conjure fully animated 3D characters, worlds, and game mechanics with nothing but voice or text prompts. While the company positions Dreamer as a kind of creative sandbox, it's the front-end of a much deeper effort to rewrite the architecture of game development. 'We've spent thirty years learning how to do it the hard way,' said Twain Martin, Upheaval's co-founder, CTO, and a veteran of Blizzard, Arcane, Obsidian, and other AAA studios. 'Now we're asking, what happens when AI does it all?' Enchanted market prompted with Upheaval. The answer, if Upheaval is right, could redefine the entire game development pipeline. Dreamer is the user-friendly gateway, built for solo creators and small teams, while the real power lies in Nexus, a back-end platform for spawning and managing AI-driven agents, which can best be described as NPCs with memory, goals, and autonomy. These agents can speak and interact with players, and build, script, and even populate game environments on their own, without any prompting at all. Unlike existing AI tools that require deep technical integrations, Nexus is designed for natural language instruction. 'You can give these agents a job with your voice,' said Aaron Moon, co-founder and Head of Product. 'We're building a world where you don't need a QA pipeline to test every change. The AI handles logic and behavior live, at runtime.' A 2D image can become a rigged, animated 3D image in seconds. FEATURED | Frase ByForbes™ Unscramble The Anagram To Reveal The Phrase Pinpoint By Linkedin Guess The Category Queens By Linkedin Crown Each Region Crossclimb By Linkedin Unlock A Trivia Ladder Upheaval's soon-to-be-released Nexus platform currently connects with with both Unity and Unreal Engine, generating rigged, animatable 3D assets stored in the cloud for drag-and-drop use. The toolstack supports character generation, dialogue behavior, item crafting, and environment logic. Its proprietary agent engine enables autonomous NPCs to interact with in-game objects, generate items with histories, and adapt behavior based on player input or scene context. One of the platform's more radical ideas is to treat the game itself as an emergent system that evolves alongside the player. 'Imagine an MMO where every NPC gets older, learns new skills, or changes appearance based on weather or combat,' said Martin. 'That's the 'Forever MMO' we're aiming for—persistent, dynamic, and built on agentic AI.' The Dreamer Portal, available now via free early access, is just the beginning. It functions as an aggregation layer atop dozens of AI services such as Meshy, Tripo, OpenAI, and others, thus allowing users to rapidly iterate without getting stuck in what Moon calls 'prompting paralysis.' The company plans to introduce a marketplace later this year, enabling creators to monetize agent-based games, assets, and experiences by selling them to other users. The broader vision goes beyond gaming. Moon says the same tools used to animate a blacksmith in a fantasy world could power simulations for scientists, educators, or even doctors. 'AI can surface patterns in massive datasets that humans would never see,' he said. 'One of our alpha testers is literally using it to help identify T-cell interactions in cancer research.' Upheaval's Nexus. Still, the implications for the gaming industry are immediate and disruptive. Moon and Martin are candid about the consequences. 'We're not pretending this won't eliminate jobs,' said Moon. 'But like Dreamweaver or the iPhone camera, it expands who gets to create.' Martin agrees: 'If ten people can now do the work of a hundred, we'll see more games—but only a few will break through.' Upheaval's founders are betting that those winners will be built with agents—not scripts. 'We're not building another asset farm,' said Moon. 'We're building a world that builds itself.'
