EAGLYS Inc. (Headquarters: Shibuya, Tokyo; President and CEO: Hiroki Imaibayashi; hereinafter "EAGLYS") has begun technical verification for implementing its Confidential AI technology into next-generation AI services being developed by Tokyo Electron Limited (hereinafter "TEL"), a global leader in semiconductor manufacturing equipment. EAGLYS aims to support TEL in building a trusted technical foundation that enables secure use of generative AI with the company's long-accumulated equipment technologies and process know-how. ※ Confidential AI is an approach that multi-layeredly ensures the confidentiality of input data, processed data, and models in AI services by combining hardware security technologies such as TEE (Trusted Execution Environment) with AI guardrails. It goes beyond simply encrypting stored data by ensuring protection even while AI is processing data, and cryptographically proves that input information is not learned by the AI. ■ Barriers to AI Adoption in Manufacturing DX As adoption of generative AI accelerates in the manufacturing industry, concerns over whether it is safe to input proprietary technical data into AI systems have become a major barrier to broader implementation. Currently, trust in AI services relies on terms of service and contractual agreements rather than technical safeguards. TEL aims to integrate its decades of accumulated semiconductor manufacturing expertise with AI to improve production efficiency and enhance customer services. To achieve this, it is essential to have technical—rather than merely contractual—assurance that sensitive data is protected. ■ EAGLYS' Approach EAGLYS provides "Confidential AI" technology that protects data even while being processed by AI. Specifically, by running AI within a hardware-isolated execution environment (TEE), it ensures that third parties, including the service provider, cannot access input data. Furthermore, through cryptographic attestation mechanisms that verify the protection envi