Slow technology rollouts and complex integrations have long hampered enterprise customer service enhancements. Oracle told a different story in its quarterly earnings call, with customer metrics indicating that the arrival of AI agents is changing the implementation model for customer service technology, making it lighter and in some cases, completely self-service.
This shift has implications for how enterprises plan, buy and scale technology in the coming years. But investors remain wary about how quickly these AI investments will pay off.
Oracle Touts New Software Implementation Baseline: Weeks, Not Years
The highly-regulated healthcare sector is rarely known for rapid software deployments. That’s why a comment from Mike Sicilia, Oracle’s co-Chief Executive Officer, stood out:
“The go lives for these clinical AI agents—a new breed of SaaS applications for us—are measured in a matter of weeks. So you’re looking at an industry like healthcare where it would take months or years to get anything done at that magnitude.”
“And by the way, the customers are implementing these things all by themselves,” Sicilia added. “They don’t need us to help them. You just roll them out and they work.”
The healthcare implementation is in its early days, as Sicilia noted that the clinical AI agent has 274 customers live in production.
But if one of the most complex sectors can move to self-service AI adoption, other industries may encounter fewer barriers than expected. For teams used to large-scale service transformations, the ability to roll out new capabilities without long cycles is a substantive shift.
AI projects often launch with optimism but lack documented improvements.
Telecom operator TIM Brasil has built AI agents on Oracle Cloud Infrastructure (OCI) targeting customer-facing workflows, including automated bill comparisons and explanations. Oracle has signed a five-year expansion of a partnership to migrate the company’s data center to OCI in 2021.
In the initial rollout of seven out of 24 projects, call center flows are managing end to end with 90 percent accuracy and the company is experiencing 15 percent fewer network failure interventions, Sicilia said. The telco has reported 30 percent faster service times for customers and 18 percent faster issue resolution, resulting in a 16 percent increase in customer satisfaction. It will bring six more projects online soon.
The pace of the rollout indicates that AI agents can increasingly handle multiple concurrent use cases.
Unified Data Access Power Smarter AI Agents
Larry Ellison, Oracle’s Chairman and Chief Technology Officer, highlighted another challenge familiar to CX teams: fragmented systems make it difficult for AI to deliver context-rich service.
Ellison pointed to three major shifts in how Oracle is positioning its data and AI strategy. First it is making the Oracle Database fully multi-cloud by embedding it inside other major clouds; second, adding vector capabilities so the Oracle Database can serve as a secure AI-ready vector store that models can reason over; and third, creating an AI Data Platform that can catalog, vectorize, and unify data across all clouds and systems so a single AI query can draw on an organization’s entire data landscape.
“By treating all of your data holistically, the combination of AI models, plus the Oracle AI Database and AI Data Platform, breaks down the walls that isolate and fragment your data… We take all of your data and unify it so you can ask a single question, and the AI models can find the answer to that question regardless of what data store it’s in.”
Regardless of platform, enterprises will need unified data environments if they intend to get full value from AI agents.
This plays into one of Oracle’s strengths, according to Rebecca Wettemann, CEO of Valoir:
“Oracle has a strong CX story that can be made even stronger when it talks about other data AI can leverage across the enterprise suite. Oracle has a real opportunity to drive differentiated value for CX customers by leveraging AI to automate processes that span beyond CX data to leverage other data Oracle has, like supply chain data that can help inform customer service’s ability to promise products or HR data that can help field service optimize service plans.”
Infrastructure Matters, Especially During Peak Service Windows
AI agent performance depends heavily on the compute layer beneath it, which Clay Magouyrk, Oracle’s co-CEO, emphasized when discussing high-traffic retail periods.
“The holiday season is a peak period for many of our retail and consumer customers. It is our responsibility at OCI to deliver the most secure, highest performance, and highest availability infrastructure to support these customers at the scale they need.”
Those customers include transportation company Uber, which has surpassed 3 million cores on OCI, powering its highest-ever traffic on Halloween and e-commerce platform Temu, which scaled to almost 1 million cores for Black Friday and Cyber Monday.
Oracle’s infrastructure business has grown by 66 percent year over year, driven by strong demand for AI compute as well as cloud natives, dedicated regions and multi-cloud. During the last quarter, the company handed over close to 400 megawatts of data center capacity to its customers, Magouyrk said.
The company has ambitious goals for capacity expansion globally, with plans to add 64 regions to the 147 live customer-facing regions it operates globally.
Oracle is seeing increasing demand for database services across all clouds, reporting a 817 percent year-over-year jump in multi-cloud database consumption. The company launched 11 multi-cloud regions during the quarter, bringing it to 45 regions live across AWS, Azure, and GCP 27, with more planned over the next month.
Sicilia also emphasized an architectural trend that will matter to teams planning multi-year roadmaps. Oracle has more than 400 AI features live in its Fusion cloud apps. infrastructure and AI data platform “allows our customers to very easily build enterprise lake houses, AI agents, and applications that leverage built-in, not bolted on… AI to transform their business.”
The industry is moving toward integrating AI into operational systems rather than layering it on top of them, which can result in inefficient, complex “Frankenstacks.” Built-in AI, whether from Oracle or any other vendor, reduces integration points and shortens the path from experimentation to production. That shift has implications for budgets, deployment strategy and governance.
Throughout the earnings call, the first for new co-CEOs Magouyrk and Sicilia, several themes emerged that matter for executive-level planning: AI agent deployment timelines are shrinking and customers are increasingly implementing these tools on their own rather than relying on external teams. Early deployments are already showing measurable gains in areas like resolution speed, customer satisfaction and accuracy. At the same time, the underlying infrastructure required to support AI-driven service is becoming a critical factor, and unified access to enterprise data is proving essential for more advanced AI use cases.
AI Momentum Fails to Erase Investor Skepticism
Despite the momentum around AI agents and accelerated deployments, the market’s response to Oracle’s earnings call showed that investors remain cautious about the broader AI landscape.
The share price has plunged by around 40 percent since its peak in September and fell further after the call.
In September, Oracle scored a $300BN, five-year contract to deliver computing power to ChatGPT-developer OpenAI. But the market is growing concerned about the exposure to OpenAI, which is facing questions around its profitability and ability to pay for its massive infrastructure deals.
“It’s not surprising Sicilia is highlighting the opportunity for AI beyond data centers given the stock price beating Oracle has been taking,” Wetteman said. “After previous earnings calls that have been largely focused on AI inferencing and data center infrastructure, and RPOs focused on the OpenAI and other yet-to-be monetized opportunities, Oracle needs to show it has a more diversified strategy.”
Oracle’s revenue fell short of expectations, rising 14 percent year on year to $16.1bn, compared with the $16.2bn analysts had projected, and its sharply higher capital spending plans for AI data centers raised questions about near-term profitability and cash flow.
That led to the share price dropping by as much as 16 percent on December 11 and renewed concerns across the tech sector about how quickly large AI investments will translate into returns. The financial community is still waiting to see how these large infrastructure bets pay off.
