Alphabet Chief Executive Sundar Pichai is steering the company deeper into enterprise software, signaling to investors and customers at Google’s annual cloud event in Las Vegas that AI agents - autonomous, human-like digital assistants - are central to its plan for turning artificial intelligence into a dependable revenue stream.
Beginning Wednesday at the three-day conference, Pichai alongside senior Google executives will present the company’s AI offerings as production-ready infrastructure tailored to enterprise buyers, a group Google and other AI leaders increasingly view as the most consistent source of industry revenue.
On Wednesday, Google announced it would consolidate several of its AI offerings under the umbrella name "Gemini Enterprise." The move notably includes a rebrand and enhancement of Vertex AI, the cloud tool that lets enterprise customers select and deploy different AI models for business use.
Along with the rebranding, Google disclosed new governance and security capabilities aimed at AI agents. These agents are designed to plan, decide, and act autonomously for users, a rapidly growing area that has prompted concern over safety, reliability, and appropriate oversight.
"There’s definitely a strategic shift as the models become much more sophisticated," Google Cloud CEO Thomas Kurian told Reuters in an interview, noting that Vertex AI’s primary use case has recently moved away from traditional machine learning workflows toward a surge of customers building bespoke AI agents.
The emphasis on agents reflects a broader industry trend. Major AI companies, including OpenAI and Anthropic, have been reallocating resources toward business customers in recent months, concentrating on downstream applications that put models to work in specialized enterprise tasks such as agent construction and coding assistance.
Kurian positioned the competitive landscape as one that will be decided by agents, governance, and large-scale enterprise rollouts rather than solely by coding tools. He suggested that some coding-related announcements would be reserved for Google’s I/O developer conference in May. "Some people are using the models to write code. They can use Gemini and also other tools like Claude," he said. "But in other cases, we have unique things. There’s capability in the platform that nobody else offers."
Google’s long-term strategy favors building a comprehensive array of internal capabilities - from models to custom chips - rather than relying primarily on third-party suppliers. Executives argue that this integrated approach, together with substantial investments in data centers and networking gear, has helped Google gain traction among enterprise customers.
Even so, Google Cloud remains behind some rivals in overall market share despite recent gains. The company’s cloud market share reached 14% at the end of 2025, though it still trails Amazon and Microsoft, according to data cited from Synergy Research.
At customer sites the shift toward AI-enabled operations is already visible. Marcia Brey, a senior executive at GE Appliances and a Google Cloud customer, told Reuters that the combination of Google’s toolset and enterprise data already hosted in Google Cloud enabled her logistics and distribution team to deploy AI more quickly than other products the company had evaluated.
Alongside traditional cloud providers and hyperscalers, a distinct category of competitors has arisen: model providers. Companies that first competed on model strength are now moving downstream, building applications and plug-ins that integrate models into existing enterprise software. Coding assistants and connectors that make models useful inside enterprise workflows have become notable early commercial channels, offering a route to revenue and a way to demonstrate returns on heavy model investments.
Where some rivals have placed their coding tools at the forefront of their enterprise pitches, Google’s cloud leadership has put more emphasis on agents and governance at this conference. Kurian said that while coding remains an important use, the company sees unique platform capabilities and an enterprise-focused value proposition centered on agents and control features.
The company also highlighted new security and governance tools for agents, reflecting rising concerns among enterprise customers about the safety, reliability, and oversight of autonomous AI systems. Those concerns, and the need to prove commercial returns from large generative AI investments, have become central challenges for the broader industry.
Key takeaways
- Google is consolidating AI tools under Gemini Enterprise and expanding Vertex AI to support custom enterprise agents.
- Executives are pitching agents, governance, and enterprise deployment as the primary arenas for monetizing AI, amid competition from model providers and hyperscalers.
- Sectors most affected include cloud infrastructure, enterprise software, and logistics operations that host and use AI for production workloads.
Risks and uncertainties
- Safety and oversight - AI agents’ autonomous capabilities raise concerns about safety, reliability, and governance, impacting enterprise adoption in regulated industries.
- Commercial returns - Pressure to demonstrate profitable returns on extensive generative AI investments creates uncertainty for cloud providers and their customers.
- Competitive pressure - Intense competition from other hyperscalers and model providers could influence market share and pricing in the cloud and enterprise AI markets.