Image: Samsung
Samsung has for the first time publicly unveiled its own generative AI model dubbed Samsung Gauss on Wednesday.
The South Korean tech giant’s generative AI model is named after famed mathematician Carl Friedrich Gauss who established the normal distribution theory, popularly known as the bell curve.
The name “reflects Samsung’s ultimate vision for the models, which is to draw from all the phenomena and knowledge in the world in order to harness the power of AI to improve the lives of consumers everywhere,” the company said during Samsung AI Forum, an annual AI forum that Samsung Research and Samsung Advanced Institute of Technology has hosted for experts and academics since 2017, in Seoul.
Samsung said the model, which was developed by its research arm Samsung Research, was currently being used on employee productivity within the company but will be expanded to product applications in the future.
Samsung Gauss consists of Samsung Gauss Language, Samsung Gauss Code, and Samsung Gauss Image.
According to the tech giant, Language is a generative language model designed to enhance work efficiency by helping in tasks such as writing emails, summarizing documents, and translating, Samsung said, and can also enhance consumer experience by providing smarter device control when applied to products. The model itself consists of various models to make it applicable on the cloud and on-device.
Code and the coding assistant called code.i that it is based on it made for in-house software development of companies. It allows developers to code easily and quickly and through an interactive interface supports functions such as code description and test case generation, Samsung said.
Image meanwhile can easily generate and edit images, allowing the application of style changes, additions, and conversion of low-resolution images to high-resolution, the tech giant added.
These models can be applied on-device to protect the private information of consumers while Samsung was also working to ensure safe AI usage through its own AI Red Team that checks security and privacy issues that can occur during data collection, AI model development, service deployment, and through the generated results.
