Yating FedGPT Wins 2027 Taiwan Excellence Award, Safeguarding Enterprise Data Sovereignty with Fully On-Premises AI

Benchmarks & AwardsPublished:2026.11.11By:AILabs PR

[Taipei, Taiwan] As generative AI becomes increasingly integrated into core business operations, enterprises face growing challenges in protecting sensitive data, managing high hardware investment costs, and ensuring AI models meet local regulatory requirements. FedGPT, an enterprise-grade Sovereign AI solution jointly developed by Yating and Taiwan AI Labs, has been recognized with the 2027 Taiwan Excellence Award. Combining fully on-premises deployment, Traditional Chinese language optimization, and low-code AI Agent workflows, FedGPT enables enterprises to adopt generative AI while maintaining control over their data, reducing hardware and operational costs, and demonstrating the innovation capabilities of Taiwan-developed AI technology.

Established by Taiwan's Ministry of Economic Affairs in 1993, the Taiwan Excellence Award is one of the country's most prestigious recognitions of industrial innovation. Each year, products are rigorously evaluated across four key dimensions: R&D, Design, Quality, and Marketing, while also meeting the requirement of being manufactured in Taiwan. The Taiwan Excellence mark represents recognition of product innovation and quality, serving as an important symbol of Taiwan's commitment to bringing high-quality products to international markets. FedGPT's recognition reflects professional acknowledgment of the team's achievements in enterprise AI research and development, product design, and practical market applications.

Benson Tu, General Manager of Yating Intelligence, said, "Enterprises should not have to compromise data security or bear excessive hardware costs to adopt AI. This is particularly important in highly regulated industries such as finance and healthcare, where organizations manage substantial amounts of sensitive information and specialized knowledge. They need AI solutions that address information security, local regulatory requirements, and real-world business needs. With FedGPT, we aim to help enterprises build their own AI Agent teams without relying on external public cloud services or investing in prohibitively expensive hardware. Receiving the Taiwan Excellence Award is an important encouragement for our team and recognizes our continued commitment to developing Taiwan's homegrown AI technologies."

Cheen Liao, Chairman of Yating Intelligence, said, "AI has become a critical foundation for global industrial competitiveness. Taiwan must not only be capable of adopting international AI technologies but also develop its own capabilities to maintain control over data, models, and computing environments. FedGPT reflects our commitment to enabling Taiwanese enterprises to benefit from AI innovation while retaining ownership of their digital assets and core knowledge. This recognition from the Taiwan Excellence Award highlights the progress of Taiwan-developed AI technologies in enterprise applications. We look forward to helping more industries establish trusted AI environments."


FedGPT: Fully On-Premises Deployment for Accessible Enterprise Sovereign AI

FedGPT adopts a fully on-premises deployment architecture, keeping enterprises' sensitive data and AI processing within their internal environments. By eliminating the need to transmit data to external public cloud services, it helps reduce the risks associated with data leakage and cross-border data transfers, while allowing organizations to manage data access and usage in accordance with their own security policies and regulatory requirements. This makes FedGPT particularly suitable for industries with stringent data privacy and information security requirements, including finance, healthcare, and government.

In terms of model capabilities, FedGPT is deeply optimized for Traditional Chinese and Taiwan-specific linguistic contexts. Its models are trained and optimized for specialized domains, including healthcare and financial regulations, enhancing their understanding of local terminology and industry knowledge. The platform also integrates Automatic Speech Recognition (ASR), multimodal Retrieval-Augmented Generation (RAG), and Vision-Language Models (VLM), supporting applications such as medical record summarization, intelligent voice-assisted clinical handovers, and multimodal enterprise knowledge bases.

Furthermore, FedGPT offers low-code workflows and AgentTeam capabilities, allowing users without programming backgrounds to combine multiple AI Agents through an intuitive node-based interface. Enterprises can build specialized AI teams that collaborate to automate complex business processes, lowering the technical barriers to AI adoption. Through model optimization, FedGPT also reduces reliance on high-end hardware. Under specific testing conditions, it can achieve energy savings of up to 94% compared with larger models, helping enterprises balance AI performance, hardware investment, operational costs, and energy efficiency.


From Yating Transcription to Enterprise Sovereign AI: Advancing Taiwan's Homegrown AI Capabilities

Yating Intelligence is a strategic partner of Taiwan AI Labs, with long-standing expertise in speech recognition, speech synthesis, information security, and generative AI. Building on the widespread adoption of its flagship Yating Transcription service, the company continues to extend its Traditional Chinese language technologies into the enterprise market, developing a diverse portfolio of speech and generative AI applications.

The 2027 Taiwan Excellence Award recognition of FedGPT not only highlights Yating Intelligence's R&D achievements in advancing from consumer-facing AI tools to enterprise-grade Sovereign AI solutions, but also demonstrates the growing potential of Taiwan-developed AI technologies in driving industrial digital transformation. Looking ahead, Yating Intelligence will continue to advance localized AI technologies and enterprise applications, empowering organizations across industries to adopt AI securely and effectively while maintaining control over their data sovereignty.

 

Tags