The Original is Back... IBM Targets Enterprise Customized AI with 'WatsonX'

WatsonX Global Launch Including Korea in July
Differentiating AI Training Data Quality and Utilization Environment

"Having worked in AI for a long time, I have always considered how companies can utilize AI. The conclusion is to expand the AI domain with trustworthy data."

Ji-eun Lee, Chief Technology Officer (CTO) of IBM Korea, said this at a media briefing held on the 13th at the International Finance Center (IFC) office in Yeouido, Seoul. The event was organized to introduce the AI platform 'WatsonX,' which is scheduled to launch in July.

Lee Ji-eun, Chief Technology Officer (CTO) of IBM Korea, introduced WatsonX at a media briefing held on the 13th at the International Finance Center (IFC) office in Yeouido, Seoul.
[Photo by IBM Korea]

WatsonX carries the meaning of being the next generation of IBM's AI model 'Watson.' Watson is an AI computer capable of answering natural language questions. It became known worldwide when it won the US quiz show 'Jeopardy' in 2011 by defeating human champions, showcasing IBM's AI technology. However, due to high costs and other factors, the related business did not grow. This is why the medical AI division 'Watson Health' was sold last year.

The differentiating factor of the revived 'WatsonX' is data. It created foundation models by selecting trustworthy and unbiased data. Foundation models are AI trained in advance on vast amounts of raw data. Since these models are already trained, companies using them can save time and costs related to preparing additional data and training. Unlike Google and Microsoft (MS), which compete in terms of the scale of training data, IBM aims to compete based on data quality. Additionally, it allows enterprise customers to apply AI in any environment. Regardless of storage location?on-premises servers, single cloud, or multi-cloud?it enables access to all data and AI-based operations.

Seok-jae Choi, General Manager of Data & AI Technology Sales at IBM Korea, emphasized, "Adopting AI has become a necessity for companies, but they still face data issues," adding, "We will provide AI foundation models wherever customer data exists."

In fact, an IDC survey revealed that data is distributed across various environments, making AI utilization difficult. The data itself is often outdated and diverse, including documents, images, and videos, so most of the time is spent organizing and integrating data.

WatsonX consists of three components: WatsonX.AI, WatsonX.Data, and WatsonX.Governance. WatsonX.AI is a kind of AI studio that provides AI foundation models. It supports companies in building, running, and deploying their own AI based on foundation models. WatsonX.Data is a data repository optimized for AI, enabling AI to quickly access and efficiently utilize data. WatsonX.Governance is a tool for trustworthy AI. IBM plans to launch WatsonX worldwide, including Korea, in July, with Korean language services scheduled for the first quarter of next year.

This content was produced with the assistance of AI translation services.