Alibaba (阿里巴巴集团) has been actively developing and deploying digital humans across a range of applications using AI and 3D technologies. These digital humans are used for e-commerce, virtual livestreaming, brand ambassadorship, customer service, and content generation. Products and services include AI-powered avatars capable of real-time interaction, multilingual communication, and personalized shopping guidance. Alibaba Cloud offers tools like LivePortrait to create talking avatar videos from a single photo and voice input. The company has open-sourced several frameworks such as MNN and MNN3DAvatar to support developers in building 3D digital humans. Notable digital human projects include "Dong Dong," a virtual spokesperson for the 2022 Beijing Winter Olympics, and “Xiao Mo,” a sign-language translating employee created with Damo Academy. Alibaba’s platforms, including Taobao and Alibaba International, provide SaaS tools and governance rules for AI livestreams and virtual hosts.
Eddie Wu (吴泳铭) is the current Chief Executive Officer of Alibaba Group and a long-time core architect of the company’s technology and platform strategy, having previously served as Chairman of Alibaba Cloud and held senior roles across core commerce and infrastructure units. His background is deeply technical, with a focus on large-scale systems, cloud computing, and AI platformization, and under his leadership Alibaba has accelerated an AI-first strategy that treats generative AI, multimodal models, and digital humans as foundational capabilities embedded across e-commerce, international platforms, and enterprise SaaS. Wu’s role is primarily strategic and integrative, aligning DAMO Academy research, Alibaba Cloud tooling, and consumer-facing products such as Taobao livestreaming so that digital humans function not as experiments but as scalable, governed commercial infrastructure.
Wang Jian (王坚) is the founder of Alibaba Cloud and one of the most influential technical figures behind Alibaba’s AI and digital human capabilities, known for establishing the cloud-native, open-source-oriented architecture that supports real-time avatars, 3D digital humans, and multimodal interaction systems. With a background in distributed computing and systems engineering, Wang championed long-term investment in foundational AI infrastructure and research-to-production pipelines, enabling technologies such as MNN, MNN3DAvatar, and avatar generation tools like LivePortrait to be deployed at scale. Although no longer managing day-to-day operations, his technical vision continues to shape how Alibaba builds developer platforms and industrial-grade digital humans that can operate reliably across commerce, customer service, and large public events.
Alibaba DAMO Academy (阿里达摩院) is the research arm of Alibaba Group and is a major internal source of the multimodal and speech-vision capabilities that underpin Alibaba’s digital-human efforts, spanning speech recognition and synthesis, talking-head and body-motion generation, video understanding, and conversational interaction that can be packaged into interactive avatars for customer service, livestreaming, and enterprise interfaces; in this context, its work typically sits at the “foundation-to-application” layer, where research prototypes in audio-driven facial animation, lip-sync and gesture alignment, and multimodal perception are translated into developer-facing components and reference implementations, often distributed through ModelScope (魔搭) and then productized as deployable capabilities via Alibaba Cloud virtual digital human services, with the practical emphasis on real-time performance, controllability, identity consistency, and safety controls needed to operate digital humans at scale across commercial scenarios.
Alibaba Cloud (阿里云) offers a virtual digital human platform integrated with its broader cloud ecosystem, enabling AI-driven avatars for use in livestreaming, customer service, and marketing. Its OpenAPI developer portal provides SDKs and tools to build and deploy interactive digital humans with real-time animation, voice, and lip-sync capabilities. These avatars are positioned as brand ambassadors or virtual assistants across multiple channels. The platform supports independent deployment, allowing businesses to create customizable AI characters based on Alibaba Cloud’s infrastructure, leveraging large language models like Tongyi Qianwen for conversational functions.
Zhang Jianfeng (张建锋), also known as Jeff Zhang, is a senior technology executive at Alibaba Group and a former President of Alibaba Cloud Intelligence, recognized as one of the principal architects of Alibaba Cloud’s technical stack and organizational structure; with deep expertise in large-scale distributed systems, cloud infrastructure, and AI platform engineering, he has overseen the development of core cloud and AI capabilities—including computing frameworks, model deployment systems, and enterprise AI services—that underpin Alibaba Cloud’s digital human, virtual agent, and conversational AI solutions, even though specific avatar or digital human products are typically implemented by specialized internal teams or external partners rather than attributed to him individually.
AutoNavi (高德软件有限公司), a subsidiary of Alibaba, has developed advanced digital human technologies integrated into its navigation and location-based services. Its core engine, HumanRig, powers 3D virtual avatars used in personalized navigation, IP voice packages, and dynamic in-app visual elements. AutoNavi has also open-sourced components of its digital human framework, focusing on audio-driven realism and interactive experiences. These avatars are designed for applications such as AR navigation and digital storytelling, aligning with Alibaba’s broader smart city and AI strategies.
Hou Jun (侯军) is identified in Chinese corporate records as the Chairman and President of AutoNavi Software Co., Ltd., the Beijing-based digital mapping, navigation, and location services provider that is a subsidiary of Alibaba Group. He oversees the company’s strategic direction, operational management, and integration of advanced technologies such as AI, digital human frameworks, and spatial intelligence into AutoNavi’s core products and services. Under his leadership, the company has expanded beyond traditional map and navigation tools toward AI-native applications and service ecosystems aligned with Alibaba’s broader smart city and intelligent spatial services strategy.
Guo Ning (郭宁) serves as Chief Executive Officer (CEO) of AutoNavi. In this capacity, he is responsible for executing the company’s product, technology, and market strategies, including the integration of advanced AI capabilities, digital characters, and navigation innovation into AutoNavi’s mobile and enterprise offerings. Guo Ning has publicly articulated AutoNavi’s transition from a navigation tool to a spatial intelligence platform, emphasizing user-centric intelligent agents and proactive services that anticipate user needs. His role is central to driving the company’s evolution within Alibaba’s technology ecosystem and maintaining its competitive position in China’s location-based services market.
PixelAI (阿里巴巴 PixelAI 团队) is an Alibaba-affiliated research and development team focused on visual computing technologies used in digital human creation and enhancement. The team is best known for developing the TaoAvatar system, a high-fidelity, real-time 3D full-body avatar solution based on 3D Gaussian Splatting. PixelAI has released several notable projects through its GitHub page, such as TaoAvatar and GaussianTalker, which support advanced AI-driven avatars with facial expressions, gestures, and real-time speech interaction. Their technologies are designed to work on mobile and AR devices, including the Apple Vision Pro. PixelAI has also developed tools for image enhancement, video restoration, real-time portrait segmentation, and AR-based product interaction (e.g., virtual try-ons), and has won awards in national broadcasting and AI competitions for its innovations in digital human and video processing technologies.
Zhiwen Chen (陈志文) is a staff algorithm engineer at Alibaba Group and the publicly identified project lead of the TaoAvatar system within the PixelAI team. His work focuses on animatable human reconstruction, neural rendering, and real-time digital human systems, with an emphasis on production-ready pipelines that can operate on consumer devices. As project lead, he is responsible for system architecture decisions, research direction, and bridging experimental avatar research with Alibaba’s applied platforms in e-commerce, AR interaction, and immersive computing.
Jianchuan Chen (陈建川) is a core PixelAI researcher and equal-contribution first author on the TaoAvatar project, indicating a primary role in algorithmic design and implementation. His research contributions center on 3D avatar reconstruction, neural representation learning, and real-time full-body digital humans using Gaussian-based rendering methods. He is also a recurring contributor across PixelAI open-source releases, including GaussianTalker, suggesting sustained involvement in both facial and full-body avatar systems.