Samsung starts development of its ChatGPT-like Large Language Model for internal use

The development of Samsung's LLM solution started this month, and it is being led by Samsung Research. The South Korean firm has reportedly invested a lot of manpower and resources in this development. GPUs are extremely important to train LLMs, and some officials claimed that all of the company's GPU resources are being used for LLM development. Samsung is planning to complete the development of the first version of its LLM within two months.

Samsung's LLM-based AI solution to be used for faster document summarization and software development

Currently, the South Korean firm is planning to use its LLM for document summarization, software development, and language translation. The company hasn't decided whether to make its AI solution available for general consumers. Initially, Samsung started using third-party generative AI solutions, including ChatGPT, but its internal data was leaked while using them. Hence, the company banned the use of third-party solutions and started the development of its own AI solution.

Samsung reportedly thinks that such AI solutions can drastically reduce the time needed for software development and semiconductor design. The company was engaged in talks with various firms, including Google, Microsoft, Naver, and OpenAI. However, in the company's top-level management meeting that was also attended by Lee Jae-yong, it was decided that Samsung should develop its own AI solution.

Various South Korean firms are developing their own LLM-based generative AI solutions, including Kakao's KoGPT, LG's ExaOne, Naver's HyperCloverX, and SKT's A-Dot. Most of these solutions can draw and write like humans. However, the accuracy of data and information privacy are among the biggest hurdles right now.