ATGL Partners with HKUST to Launch AlphaMind Lab
On the afternoon of March 27, 2025, The Hong Kong University of Science and Technology (HKUST)and Nasdaq-listed Alpha Technology Group Limited (Nasdaq:ATGL) held a signing ceremony atHKUST. The two parties announced the establishment of the"AlphaMind Lab," an AI research institute focused on developing the revolutionary "Alpha Engine." This initiative aims to overcome the traditional AI development bottleneck related to high dependence on vast quantities of high-quality data and manual annotation, thus promoting the inclusive advancement of AI technology. The collaboration is supported by ATGL's donation, highlighting the strong connection between local technology firms and academia in the core field of AI.
Signing Ceremony: Industry-university collaboration promotes AI underlying technology innovation
The signing ceremony was co-hosted by Professor Yang WANG, Vice-President for Institutional Advancement of HKUST, and Mr. Anthony TSANG, Executive Director and President of ATGL. The HKUST delegation included Associate Professor Can YANG from the Department of Mathematics, Assistant Professor Yangqiu SONG from the Department of Computer Science and Engineering, and Manager Jacqui SHI from the Development and Alumni Office. ATGL's senior management was represented by CEO Sean Leung, CFO Larry CHOI, CCO Jerry Tsang and Vice President Lucas TSANG.In his speech, Professor Wang emphasized, "The technical concept of Alpha Engine directly addresses the pain points in AI development—data and computing power thresholds. HKUST's research expertise in machine learning and algorithm optimization will complement ATGL's industrial experience, accelerating the transition of technology from the lab to the market."Alpha Engine: Empowering SMEs to Overcome Data Limitations
According to the institute, the primary goal of the Alpha Engine is to drastically reduce the time and cost associated with high-quality data collection and manual annotation by utilizing the "training AI with AI" autonomous learning architecture. This innovation aims to compress the time required to develop a dedicated AI model from the traditional 6-9 months to just 3 days or even 3 hours.
Anthony pointed out, "The current AI ecosystem relies on three pillars—GPU computing power, high-quality data and models, and engines. Small and medium-sized enterprises struggle to compete with larger firms due to limited data resources. If Alpha Engine is successfully developed, it will redefine the landscape, making AI development independent of data scale."
Since 2017, ATGL has been dedicated to the customized application of large language models (LLM), serving various government departments and leading enterprises. Anthony highlighted that the company has outperformed international technology giants in bids due to its technical accuracy, early data accumulation, and cost advantages. However, the AI transformation needs of SMEs remain unmet.
The breakthrough of Alpha Engine aims to turn customized AI from a "luxury" into a "necessity." The company is determined to become the "NVIDIA" of AI applications and bring transformative changes to the global AI ecosystem, aiming to empower all companies to utilize their own exclusive AI.
Reporter Q&A: Vision and Industry Impact
Anthony Tsang explained why his company chose to collaborate with HKUST. He replied, "HKUST's ground-breaking work in enhanced learning and unsupervised learning offers a robust theoretical basis for designing Alpha Engine’s algorithms. ATGL's industry data and engineering expertise enable us to convert these academic advancements into practical applications." In addition, Larry Choi, ATGL's CFO, remarked, "ATGL intends to create a fund to maintain our partnerships with local universities and colleges in AI research. This initiative will also support Hong Kong’s primary and secondary school teachers by funding skill development in using AI for administrative and educational purposes, thus hastening the growth of AI in Hong Kong’s education sector."
Regarding the anticipated impact of the technology, Professor Can Yang added, "If Alpha Engine meets its targets, the efficiency of AI development could increase by more than 100 times. Data-sensitive industries, such as healthcare and manufacturing, will be able to deploy models quickly while bypassing privacy barriers, enabling SMEs with limited resources to optimize operations through cost-effective customized AI."
