AWS chief executive officer Adam Selipsky [Courtesy of AWS]

<이미지를 클릭하시면 크게 보실 수 있습니다>

Amazon Web Services Inc. (AWS), the world’s largest provider of cloud services, is entering the generative artificial intelligence (AI)-based multi-modal large language model (LLM) market, a move that will help enhance its presence in the generative AI market that is dominated by its rivals Google LLC and Microsoft Corp.

AWS released the multi-model version of its LLM, Titan, on Wednesday at its annual re:Invent conference held in Las Vegas, the United States.

Multi-modal LLM refers to generative AI models that can recognize and generate various forms of data, including natural language text, images, voice, and videos.

AWS also debuted Titan Image Generator, a newly developed AI image synthesis model based on Titan.

“(You) can use the model to easily swap out an existing (image) background to a background of a rainforest,” according to Swami Sivasubramanian, vice president for data and machine learning services at AWS. “(You) can use the model to seamlessly swap out backgrounds to generate lifestyle images, all while retaining the main subject of the image and to create a few more options.”

AWS’s move to announce the multi-model version of Titan about seven months after unveiling Titan in April 2023 comes as multi-model LLMs have recently been recognized as a new game-changer in the global AI market.

Multi-model AI models, which are unaffected by data formats, are essential for realizing next-generation, general-purpose artificial general intelligence (AGI) that is close to human intelligence.

OpenAI already provides ChatGPT as a multi-modal model and Google is also developing the successor multi-modal model Gemini, following the LLM PaLM 2 with image input capabilities.

AWS is making a strong drive by announcing new services related to generative AI at re:Invent.

AWS accounts for the largest market share in the global cloud market but analyses suggest that it may have felt a sense of crisis with the emergence of competitors in recent years.

On Tuesday, AWS also unveiled Amazon Q, a corporate generative AI chatbot that assists in tasks such as document summarization, data generation, and code writing through natural language text communication.

Currently, Amazon Q has been officially released on the customer service management platform Amazon Connect and will be introduced into corporate solutions such as Amazon QuickSight and Amazon Supply Chain moving forward.

Unlike Google and Microsoft, who released general-purpose AI chatbots for the public first, AWS, which focused on corporate AI from the beginning, has emphasized its differentiated security.

Amazon Q accurately understands the user’s identity and authority and provides personalized services accordingly, AWS chief executive officer Adam Selipsky said. He added that Amazon Q does not use business customer content for learning purposes.

Amazon Q is priced at a lower cost compared to Google and Microsoft, ranging from $20 to $25 per person per month.