Tencent Pushes AI for Science Into University Research and Teaching
At the smart education session of the 2026 Tencent Global Digital Ecosystem Summit on 29 October, Tencent Cloud made AI for Science the centrepiece of its education pitch, arguing that a new research paradigm is arriving in universities and schools and that its cloud, model and agent stack is what will carry it there.
Shi Mei, vice president of Tencent Cloud and head of Tencent Education, said AI agents are shifting from personal productivity tools to productivity tools for enterprises and organisations, reaching into campus management, teaching and research. She said AI for Science opens up the boundaries of scientific exploration through human-AI collaboration, accelerating discovery and industrial application. Tencent already runs AI for Science collaborations with Tsinghua University, the University of the Chinese Academy of Sciences, Fudan University, the Southern University of Science and Technology and City University of Hong Kong.
The "fifth revolution" framing came from Du Yu, author of "AI for Science" and dean of the Weikezhi AI Research Institute and a former investment vice president at Sequoia Capital China, who described research moving from empirical trial and error through theoretical derivation, computational simulation and data mining to intelligence-driven discovery, with heavy investment now flowing into biopharma, materials science, meteorology and the environment.
Yang Hui, vice president of Tencent Cloud and head of Tencent Education solutions, set out the stack: an AI Infra compute base spanning training and inference, the TI-ONE model development platform and the TEFS scientific computing platform, with the WorkBuddy open ecosystem connecting research applications and expert capability into a Dayan AI4S community. On infrastructure, Tencent Cloud is providing research computing for City University of Hong Kong’s SciencePal multi-agent research assistant platform, the Shenzhen Institute of Artificial Intelligence and Robotics for Society and a biology team at the CAS Institute of Zoology. TEFS ships more than 30 pre-installed open-source packages for materials, chemistry and physics and is used by the UCAS School of Materials Science and Fudan’s School of Life Sciences.
Through WorkBuddy, research capability is being packaged as callable Buddy apps: ChinaScope’s ScienceOra decision platform launched with more than 30 research expert agents and skills covering literature analysis, hypothesis generation, scheme design, modelling and simulation, validation and results presentation; Emotibot wrapped its PyTRIO and SwanLab platforms into a model-training expert that users can iterate with conversationally; and a CNKI basic-education Buddy app added 15 skills for school teachers’ research, teaching and management. Tencent also positioned Agent Infra as a new layer supporting education companies such as TAL, with TokenHub letting enterprises route between Hunyuan and other mainstream models.
A panel with Emotibot, Quanyu Intelligence, Shanghai Jiao Tong University’s AI new-materials research centre, ScienceOra, the FlagOS open-source community and SpeedAI focused on AutoResearch, concluding that the hard part has moved from automating single steps to automating the whole chain — literature review, hypothesis, design, simulation, validation and publication — and that lowering the barrier for scientists depends on an open heterogeneous-computing ecosystem.
On adoption, Tencent said WorkBuddy has entered 100 universities across the mainland, Hong Kong and Macao since the 2026 school term began. Chongqing University is the first in the country to put WorkBuddy into a compulsory AI general-education course; Shenzhen University is building AI-plus-liberal-arts programmes with Tencent Cloud; and Dalian University of Technology’s "Lianxiaoli" campus-wide AI companion is now standard issue for more than 6,000 freshmen. The UNESCO International Centre for Higher Education Innovation, which says it has served 400,000 higher education professionals in many countries, is co-building an AI education system with Tencent and its partners. Three Tencent Education awards went to AI4S research, WorkBuddy education scenarios and AI talent development.
Shi Mei, vice president of Tencent Cloud and head of Tencent Education, said AI agents are shifting from personal productivity tools to productivity tools for enterprises and organisations, reaching into campus management, teaching and research. She said AI for Science opens up the boundaries of scientific exploration through human-AI collaboration, accelerating discovery and industrial application. Tencent already runs AI for Science collaborations with Tsinghua University, the University of the Chinese Academy of Sciences, Fudan University, the Southern University of Science and Technology and City University of Hong Kong.
The "fifth revolution" framing came from Du Yu, author of "AI for Science" and dean of the Weikezhi AI Research Institute and a former investment vice president at Sequoia Capital China, who described research moving from empirical trial and error through theoretical derivation, computational simulation and data mining to intelligence-driven discovery, with heavy investment now flowing into biopharma, materials science, meteorology and the environment.
Yang Hui, vice president of Tencent Cloud and head of Tencent Education solutions, set out the stack: an AI Infra compute base spanning training and inference, the TI-ONE model development platform and the TEFS scientific computing platform, with the WorkBuddy open ecosystem connecting research applications and expert capability into a Dayan AI4S community. On infrastructure, Tencent Cloud is providing research computing for City University of Hong Kong’s SciencePal multi-agent research assistant platform, the Shenzhen Institute of Artificial Intelligence and Robotics for Society and a biology team at the CAS Institute of Zoology. TEFS ships more than 30 pre-installed open-source packages for materials, chemistry and physics and is used by the UCAS School of Materials Science and Fudan’s School of Life Sciences.
Through WorkBuddy, research capability is being packaged as callable Buddy apps: ChinaScope’s ScienceOra decision platform launched with more than 30 research expert agents and skills covering literature analysis, hypothesis generation, scheme design, modelling and simulation, validation and results presentation; Emotibot wrapped its PyTRIO and SwanLab platforms into a model-training expert that users can iterate with conversationally; and a CNKI basic-education Buddy app added 15 skills for school teachers’ research, teaching and management. Tencent also positioned Agent Infra as a new layer supporting education companies such as TAL, with TokenHub letting enterprises route between Hunyuan and other mainstream models.
A panel with Emotibot, Quanyu Intelligence, Shanghai Jiao Tong University’s AI new-materials research centre, ScienceOra, the FlagOS open-source community and SpeedAI focused on AutoResearch, concluding that the hard part has moved from automating single steps to automating the whole chain — literature review, hypothesis, design, simulation, validation and publication — and that lowering the barrier for scientists depends on an open heterogeneous-computing ecosystem.
On adoption, Tencent said WorkBuddy has entered 100 universities across the mainland, Hong Kong and Macao since the 2026 school term began. Chongqing University is the first in the country to put WorkBuddy into a compulsory AI general-education course; Shenzhen University is building AI-plus-liberal-arts programmes with Tencent Cloud; and Dalian University of Technology’s "Lianxiaoli" campus-wide AI companion is now standard issue for more than 6,000 freshmen. The UNESCO International Centre for Higher Education Innovation, which says it has served 400,000 higher education professionals in many countries, is co-building an AI education system with Tencent and its partners. Three Tencent Education awards went to AI4S research, WorkBuddy education scenarios and AI talent development.
JMDedu is a leading B2B EdTech Media in China that informs and connects business professionals through our reports, events, and information services. Our mission is to drive the advancement of education. Our network spans the education industry in China as well as globally, including entrepreneurs, investors, startups, educators, business decision-makers, and policy decision-makers.
Copyright © 2014-2026 en.jmdedu.com All Rights Reserved