Amazon is making a significant investment in training a large language model (LLM) codenamed "Olympus," joining the likes of OpenAI and Alphabet's Google towards an AI-driven industry.
Rohit Prasad, the former head of Alexa, will be spearheading the Olympus project and will report directly to Amazon's CEO Andy Jassy.
As the head scientist of artificial general intelligence (AGI) at Amazon, Prasad assembled a team of researchers, including those with expertise in Alexa AI and the Amazon science team.
This collaborative effort aims to consolidate AI initiatives across the company, providing dedicated resources for training models.
The Olympus model, with over two trillion parameters, makes for a notable milestone for the company, potentially surpassing OpenAI's GPT-4, which boasts one trillion parameters and is considered one of the best models available.
The significance of this investment lies in the belief that having in-house developed models will enhance Amazon's offerings on its cloud computing platform, Amazon Web Services (AWS).
This move aligns with the growing demand from enterprise clients on AWS seeking access to high-performing models.
Amazon's Previous AI Efforts
Olympus is not the first time the eCommerce giant has delved into the advantages of AI.
In August, the company introduced generative AI to enhance customer experience by providing summarized customer reviews of certain products.
Two months earlier, Amazon invested $100 million to help customers create and implement AI solutions and launched its first Generative AI Innovation Center.
While the tech giant has already trained smaller models like Titan and collaborated with AI startups such as Anthropic and AI21 Labs, the Olympus project represents a substantial leap forward in scale and capability.
The company anticipates that investing in larger AI models will contribute to the attractiveness of its services on AWS.
While no specific timelines have been released for the Olympus model, Amazon revealed in an April earnings call its commitment to increasing investment in generative AI, signaling the company's strategic shift towards advancing AI technologies.
Edited by Nikola Djuric