Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
You might think that Amazon’s biggest turnaround in the AI race was its $8 billion investment in Anthropic. But AWS has also built internal core models, new chips, massive data centers, and proxies aimed at keeping enterprise customers confined within its ecosystem. The company believes these offerings will give it an advantage as companies of all shapes and sizes deploy AI in the real world.
WIRED sat down with AWS CEO Matt Garman ahead of the company’s annual re:Invent conference in Las Vegas to discuss his vision for artificial intelligence, and how he plans to expand Amazon’s lead in the cloud market over its fast-rising rivals, Microsoft and Google.
Jarman is betting that AI is a service that AWS can provide at a lower cost and more reliably than its competitors. Through Bedrock, Amazon’s platform for building AI applications, he says customers can access a variety of AI foundation models while maintaining the familiar data controls and layers of security and reliability that AWS is known for. If this idea holds up, it could help AWS dominate in the age of artificial intelligence.
“Two years ago, people were building AI apps. Now, people are building AI-driven apps in “AI is becoming a feature within large products rather than a standalone experience,” Jarman said. “This is the platform we’ve built, and this is where I think AWS is really starting to take the lead.”
Many of the announcements at this year’s re:Invent fall along these lines. Amazon has unveiled new, cost-effective AI models in its Nova series; Agents who can work independently on software development and cybersecurity tasks; Plus a new offering, Forge, allows companies to inexpensively train AI models on their own data.
The stakes are high for AWS to get this right. While Amazon’s cloud unit has dominated the smartphone era, smaller competitors like Google Cloud and Microsoft Azure have grown at higher rates since ChatGPT’s arrival. Microsoft and Google have emerged through close integration with leading AI models — the technology underpinning ChatGPT and Gemini, respectively — attracting companies eager to experiment with cutting-edge capabilities.
This rise of AWS competitors has raised questions about Amazon’s broader AI strategy, and how the incumbent will perform in the coming years.
Jarman says he’s been hearing these concerns for years, but less so in recent months. He argues that the tide is starting to turn, pointing to the company’s stronger-than-expected AWS results in the company’s third quarter as evidence that his strategy is working.