IBM supports the optimization of corporate AI adoption based on trusted data.

On the 13th, IBM introduced the 'IBM Watson X' product, scheduled for global launch including Korea this coming July, and announced its business strategy at the IBM Korea office in the IFC in Yeouido.

At the event, IBM asserted, “Trusted data is the core of enterprise AI. IBM provides reliable data and solutions utilizing Foundation models and generative AI to enable its use. Furthermore, we differentiate ourselves from competitors by supporting data integration across various hybrid environments, including on-premises servers and multi-cloud, rather than relying on a single cloud.”

'IBM Watson X' is an enterprise AI and data platform that can build and improve existing machine learning and modern foundation models.

Here, a foundation model refers to an AI model that has been trained on a vast amount of data in advance. When utilizing AI within an organization, using foundation models can reduce AI training time and costs.

'IBM Watson X' consists of 'IBM Watson X.ai,' an AI studio that helps build, execute, and deploy enterprise AI; 'IBM Watson X.Data,' a data repository; and 'IBM Watson X.Governance,' which supports policies and decision-making. The company plans to target global and domestic markets with its data and AI products this coming July. The Korean version will be released in the first quarter of next year.

Lee Ji-eun, CTO of IBM Korea, cited reliability as the core of Watson X. “IBM does not handle just any data. From the moment we create the foundation model, we base fine-tuning solely on clear data,” she said, adding, “We can also grant a degree of freedom depending on customer requirements. By partnering with HuggingFace, we enable the use of open-source data while establishing safeguards to support companies in using it freely.”
'Watson x.Data', which was the focus of introduction on this day, is a concept that combines the advantages of data lakes and data warehouses. It is a solution that simultaneously provides the flexibility of a data lake and the high-performance processing technology of a data warehouse.

Choi Seok-jae, Managing Director of Data & AI Technology Sales at IBM Korea, said, “IBM’s data lakehouse aims to combine low storage costs with high-performance computing engines.”

Watson x.Data enables cost and performance optimization tailored to the customer. It includes both structured and unstructured data, and reduces computing and storage space according to data usage, thereby reducing existing data warehouse costs by up to 50%.

In addition, Watson x.data emerged from the question of how enterprises could integrate the vast amount of data generated in the cloud with existing on-premises systems.

Executive Director Choi said, “In the early days, most lakehouses were skewed toward cloud-based systems, but IBM’s Watson x.data supports all forms of deployment environments. In other words, it is easy to access all data from a point of access, regardless of where the data is stored.”

Watson X.Data utilizes Foundation Models and Generative AI, just like other Watson X platform solutions.

For example, a representative feature is semantic automation, which enables users to utilize necessary data using natural language sentences and questions. If you say, "Add residence data to this table," just as you would converse with a chatbot like ChatGPT, the semantic automation AI performs the task. Semantic automation technology is scheduled to be officially released in early 2024.

Meanwhile, IBM continues active collaborations in the medical and space sectors. IBM is currently developing a foundation model based on medical data with Moderna. In addition, it is expected to be applied in various fields through NASA and Earth science data.

The CTO said, “The focus is on how companies apply AI, so rather than being biased toward specific fields, we will provide broad assistance through the foundation model in any domain where data exists.”