October 15, 2025, Taipei－APMIC, A global pioneer in delivering enterprise-grade AI solutions, announced today (15) that its Traditional Chinese language model, ACE-1-24B, has officially passed the Trustworthy AI evaluation conducted by the AIEC (Artificial Intelligence Evaluation Center) under the guidance of the Ministry of Digital Affairs, and has obtained formal certification reports. ACE-1-24B is among the first Traditional Chinese models in Taiwan to pass this evaluation. APMIC will continue to center on its self-developed models to deliver compliant AI models that meet evaluation and industry certification requirements, lead industry-wide model standardization, and help high-compliance sectors such as government, finance, and manufacturing accelerate the adoption of trustworthy AI—laying the foundation for the market’s shift from a “technology race” to a “trust race.”

APMIC ACE-1-24B Passes AIEC Evaluation, Advancing AI Compliance and Trust Standards for Local Industries

According to the World Trade Organization, while global enterprise AI adoption surged by 115% between 2023 and 2024, only 62% of business leaders and 52% of employees believe AI is being “responsibly deployed” within organizations. Amid this trust gap, establishing trustworthy AI models and standards has become an urgent global priority. To strengthen Taiwan’s competitiveness in the global AI landscape, the AIEC—guided by the Ministry of Digital Affairs and aligned with international frameworks (including NIST, ISO, and EU regulations)—was established to promote a locally governed and verifiable testing system covering security, fairness, explainability, privacy, and cybersecurity.

APMIC’s self-developed Traditional Chinese language model, ACE-1-24B, is among the first to pass AIEC testing. Beyond meeting international-grade verification standards, it demonstrates practical capabilities in privacy, security, and governance—marking a critical step toward AI model standardization for Taiwan’s industries.

APMIC Founder and CEO Jerry stated: “The future of AI models is no longer a race for compute, but a race for standards and trust. Only through impartial and verifiable evaluation mechanisms can enterprises confidently adopt AI and truly realize its value. ACE-1-24B’s AIEC certification signifies that Traditional Chinese models have entered a standardized path that is governable and deployable. APMIC will continue to promote clear compliance standards across industries, enabling government, finance, and manufacturing sectors to adopt AI with confidence and transform it into a core driver of digital transformation.”

ACE-1-24B Aligns with Responsible AI Trends, Enabling Verifiable and Deployable AI Capabilities for Enterprises

Stanford University’s AI Index Report 2025 indicates that among organizations that have implemented Responsible AI policies—such as cybersecurity, regulatory compliance, and privacy—over 40% report improved operational efficiency, reduced costs, and enhanced customer trust and brand reputation. This highlights compliance and responsibility as emerging core competitive advantages in AI adoption.

To address these needs, APMIC’s ACE-1-24B features a 65K long-context capability and is trained across three key domains—mathematical reasoning, everyday reasoning, and tool instructions—fully supporting Chain of Thought to enhance explainability and reliability. Through the Model Context Protocol (MCP), ACE-1-24B can connect to external systems to enable natural language–driven information retrieval, workflow operations, and decision support, creating a new intelligent human–machine interaction interface. It also supports integration with external APIs, ERP systems, and IoT devices.

Optimized specifically for high-compliance Traditional Chinese use cases in finance, government, and manufacturing, ACE-1-24B will further evolve into industry-specialized expert models. These can be efficiently deployed on-premises to ensure data sovereignty while supporting applications such as regulatory queries, technical document processing, and financial analysis—helping enterprises build a trustworthy AI brain.

APMIC (Accelerate Private Machine Intelligence Company) is a leading provider of enterprise-grade AI solutions, founded in Taiwan in 2017. Specializing in natural language understanding (NLU) technology, APMIC offers localized large language models and solutions that allow businesses to develop their own AI capabilities without programming. With applications in over 1100 enterprises across various industries and government departments, APMIC is dedicated to equipping every company with the ability to develop its own AI capabilities, to empowering everyone to seamlessly be part of the AI revolution.

For more information, visit APMIC's official website: https://www.apmic.ai/