Samsung Advances AI Strategy with Agentic Tools and Enhanced Gauss Models
Samsung is making significant strides in artificial intelligence development, enhancing its Gauss generative AI models and creating proprietary Agentic AI tools as part of its broader strategy for technological self-reliance in the competitive AI landscape.
Agentic AI Builder and Enhanced Gauss Models
Samsung Research, the company's advanced research division, has developed an 'Agentic Builder' production tool utilizing the latest iterations of its large language models. The tool leverages Gauss 2.3, Gauss 2.3 Think, and Gauss O Flash models to create a no-code, interface-based system for developing business-critical AI agents.
The Agent Builder enables rapid development of AI agents through a drag-and-drop interface that combines various components like input/output windows, AI models, and DoXA logic structures. DoXA represents Samsung's internal document analysis engine, which allows AI agents to understand complex business contexts when integrated into the system.
Sirius Knowledge Search Service
Leveraging Gauss' multimodal capabilities, Samsung has developed Sirius, an in-house knowledge search service that focuses on knowledge graph systems to overcome limitations of conventional search methods. The system can process and retrieve information across multiple formats including text, numbers, tables, images, and attachments.
Currently, Sirius is being provided as a beta service for employees, primarily used for searching product development-related knowledge and exploring specific technical tasks and information about responsible personnel. The service represents a significant advancement in enterprise knowledge management through AI-powered retrieval systems.
Advanced Image Generation Capabilities
Samsung has also developed improved image generation AI models that address common limitations in existing systems. The new models can generate images more accurately according to user requests and create representations of objects they haven't explicitly learned through training.
The company implemented a structure that incorporates reference images, allowing the AI to transform objects based on natural language prompts while preserving key characteristics without distortion. Internal usage of the enhanced image generation model has reportedly increased by 153% since the release of the newer version, indicating strong adoption and utility within the organization.
Future Integration and Product Roadmap
Samsung aims to deploy its improved Gauss AI models both internally and in future consumer products. The Galaxy S26 is expected to be the first Samsung smartphone to support Agentic AI throughout the system, potentially incorporating multiple AI models including Gauss alongside other industry solutions.
The development of proprietary Agentic AI tools and enhanced Gauss models represents Samsung's commitment to reducing dependence on external AI providers and establishing a competitive edge in the rapidly evolving artificial intelligence sector. These advancements position the company to offer more integrated and sophisticated AI experiences across its product ecosystem.
Conclusion
Samsung's development of Agentic AI tools and improvements to its Gauss AI models demonstrate the company's serious commitment to establishing technological independence in the AI domain. The creation of practical tools like the Agentic Builder and Sirius knowledge search service, combined with enhanced image generation capabilities, shows a focus on both internal efficiency and future consumer applications. As these technologies mature and potentially integrate into upcoming Galaxy devices, Samsung appears positioned to compete more effectively in the global AI landscape while reducing reliance on third-party AI providers.