Big Red gets cozy with Google to offer TPUs to AI customers

Oracle plans to add those chips alongside new Nvidia hardware and OpenAI services as businesses await AI payoffs

Oracle Cloud Infrastructure (OCI) customers will soon have a raft of new choices. Big Red is hosting an event in Las Vegas this week to inject more air into the AI bubble explain its latest AI investments, and front and center of those efforts are an expanded partnership with OpenAI and plans to diversify its chip offerings beyond Nvidia by bringing Google's Tensor Processing Units (TPUs) to OCI.

Oracle says its deal to add Google TPUs to OCI will make it the first hyperscale cloud platform outside Google's own cloud services to offer the chips. Those chips will be available for selection within OCI beginning in 2027, with capacity increasing over the course of several years.

While not mentioning the quantity of Google TPUs that will be available next year or beyond, it does seem that Oracle is attempting to diversify away from being so heavily reliant on Nvidia, with whom it has inked deals worth tens of billions of dollars to outfit its datacenters, including the reportedly troubled Texas project it’s planning with the GPU maker and OpenAI. 

REG AD

“Different AI workloads have different compute needs,” said OCI president Mahesh Thiagarajan. “Bringing Google TPUs to OCI gives customers … the flexibility to choose the right infrastructure for the performance, throughput, and economics of each AI workload.”

REG AD

Oracle isn't moving away from its partnership with OpenAI, either, announcing new ChatGPT integrations on Tuesday alongside two products that incorporate models from the House of Altman.

The first of those, Oracle AI Workspace, is billed as a “single user experience that brings OpenAI frontier models together with Oracle applications, enterprise data, and cloud services” through ChatGPT plugins that basically turn Oracle’s existing business platforms into ChatGPT-accessible resources. 

“The Oracle AI Workspace will make the breadth of Oracle’s capabilities agent-ready and accessible to ChatGPT users,” the company explained. “People can describe business goals and these plugins in ChatGPT will bring together the right Oracle tools and information to help them accomplish it.”

Along with giving OpenAI-powered tools access to selected business data, Oracle is hoping customers will also trust the models to help handle cybersecurity with the second new product, Oracle AI Security Mesh, "a managed platform that combines frontier AI models for cybersecurity … with Oracle's proven security expertise."

As you might assume from its name and description, Oracle AI Security Mesh will use OpenAI cybersecurity models to find vulnerabilities, test whether they can be exploited, and recommend fixes in areas of a company's tech stack that it authorizes the Oracle-managed service to examine.

Oracle also announced a deal with Snowflake to bring its AI data cloud to OCI as a native service. The new partnership will allow Oracle customers to purchase and use Snowflake directly from Oracle, which will handle all pricing, discounts, and billing. Oracle Universal Credits will also be usable through Snowflake on OCI, allowing users who already spend money on Oracle credits to use them for the newly available offering. 

Nvidia not exempt from announcements

Oracle may be diversifying its OCI chip offerings, but it still announced new Nvidia hardware in Las Vegas, too. 

REG AD

Big Red announced the general availability of Nvidia Vera Rubin NVL72 systems and Nvidia Vera CPUs on Tuesday, along with plans to introduce HGX Rubin NVL8 systems, though it did not specify when those would become available.

“By combining NVIDIA Vera Rubin GPUs and Vera CPUs with Oracle Acceleron networking and AI-optimized data centers, we’re helping customers keep GPUs productive, secure, and scale efficiently as their needs grow,” Thiagarajan said. 

Oracle posted record earnings in the quarter ending August 31, with revenue up 30 percent year over year as its AI investments have begun to pay dividends. Those dividends appear to be paying off for other AI firms as well, though there's little evidence of similar returns outside the tech sector, raising fresh concerns about an AI bubble that's starting to look a bit taut.

There are already signs that the AI bubble has begun to burst in China, and Oracle’s Las Vegas announcements largely involve more circular spending among AI’s top players as businesses continue to wait for results. ®