Amazon is doubling down on its AI efforts with a new group dedicated to agentic AI, aiming to help users automate tasks seamlessly without constant prompting. Part of Amazon’s commitment to advancing AI automation, this group will be led by Amazon Web Services (AWS) Vice President Swami Sivasubramanian.
“Agentic systems offer possibilities that extend far beyond today’s chatbots and will drive efficiency like we haven’t seen before,” said Sivasubramanian, who most recently was in charge of Amazon Web Services’ database, analytics and AI services, on LinkedIn. “They will orchestrate complex workflows and solve problems with human-like reasoning, while maximizing performance and cost effectiveness at scale.”
READ: Amazon unveils plans for AI supercomputer with Apple, Anthropic as first users (December 4, 2024)
AWS CEO Matt Garman added, “Agentic AI has the potential to be the next multi-billion business for AWS. We have the opportunity to help our customers innovate even faster and unlock more possibilities.”
Customers have begun adopting agent software through AWS. Amazon is also a big user internally. For instance, the company’s programmers have begun directing AWS’ Q developer service to write or update source code. AWS Senior Vice President Peter DeSantis also announced further reorganizations within AWS, consolidating AI groups Bedrock and SageMaker under the compute division and forming a new unit focused on customer experience and commerce.
READ: Amazon seeks green card allocation to help immigrant employees (August 10, 2022)
Recently, Amazon has been going all in on AI with over $100 billion investment. Amazon CEO Andy Jassy said during Amazon’s fourth-quarter earnings call on March 6 that Q4 2024’s capex spending of $26.3 billion “is reasonably representative” of what to expect on an annualized basis in 2025. For four quarters, the revenue can be estimated to be about $105.2 billion.
AWS has also recently announced plans for an AI supercomputer called the “Ultracluster,” made up of hundreds of thousands of in-house Trainium chips and a new server. Apple and Anthropic are expected to be its first users.