advertisement
Facebook
X
LinkedIn
WhatsApp
Reddit

Samsung reveals its coding, writing and image creation AI called Gauss

  • Samsung joins a growing string of companies launching AI models.
  • Currently an internal product, Gauss will eventually find its way into products and services.
  • Gauss is comprised of three models namely language, code and images.

Most, if not all Samsung users know the despair that follows when they accidentally launch the manufacturer’s digital assistant, Bixby. The assistant never really reached the same glorious heights as Apple’s Siri, Amazon’s Alexa, or Google’s poorly named, Google Assistant.

Despite this, Samsung is leaning into artificial intelligence (AI) with its own AI model, Samsung Gauss.

Announced at the Samsung AI Forum 2023 happening in Seoul, South Korea, this week, the AI model has three different products tied to it namely, Samsung Gauss Language, Samsung Gauss Code and Samsung Gauss Image.

Gauss Language is the model that resembles ChatGPT in that it’s an AI-powered chatbot you can use to compose emails, summarise documents and translate text with.

Gauss Code works in tandem with with coding assistant code.i. This solution is optimised to allow developers to code quickly and easily.

Gauss Image is a generative image model that can create and edit images. Interestingly, Samsung says Gauss Image can convert low-res images into high-res images. This may be rather useful, depending on how good it is.

Why Gauss? Samsung says the model was named after Carl Friedrich Gauss, who established normal distribution theory, a mathematical concept that forms the backbone of machine learning and AI.

Unfortunately, Gauss is only being used internally by Samsung, but a press release states that the models “will be expanded to a variety of Samsung product applications to provide new user experience in the near future”.

According to ZDNet, the models Samsung presented can be on-device as well, to protect a customer’s private information but we do wonder how these models fare on devices that don’t have the same computing power as datacenters do.

advertisement

About Author

advertisement

Related News

advertisement