Google Cloud expands Gemini enterprise agent controls

Google Cloud has expanded the general availability of features in its Gemini Enterprise Agent Platform, covering agent runtime, identity, governance and monitoring.

The additions are aimed at businesses building AI agents for longer-running tasks and more tightly controlled internal use. Newly available functions include Agent Memory Bank, Agent Runtime, Agent Identity, Agent Gateway, Agent Registry, Agent Evaluation and Agent Observability.

The update reflects a broader market shift from simple conversational assistants to software agents that can carry out multi-step work over extended periods. Businesses are increasingly testing AI systems for tasks such as sales prospecting, compliance monitoring and IT incident response, while also trying to control what those systems can access and how their actions are tracked.

Under the release, Agent Runtime allows agents to run continuously for up to seven days, supporting workflows that span several days rather than a single session. Agent Memory Bank is designed to preserve structured context from previous interactions, allowing agents to retain user preferences, prior decisions and account history.

These tools address a central technical challenge in enterprise AI deployments: maintaining continuity across work that unfolds over multiple steps, channels or handovers. Rather than requiring users to restart each interaction from scratch, the system is intended to carry context forward.

Security focus

Security and governance are also central to the update. Agent Identity introduces what Google Cloud described as a native identity and access management type for agents, with permissions tied directly to the runtime and built around a least-privilege model.

That matters because many organisations have long struggled with service accounts, dormant credentials and unclear ownership of automated systems. As AI agents take on more autonomous tasks, those identity issues become harder to ignore, particularly in regulated industries and large companies with complex internal systems.

Agent Gateway is intended to serve as a control point across an organisation's agent estate, allowing teams to apply access controls and other policy rules. Agent Registry, meanwhile, is positioned as a central catalogue for agents, servers and connections, giving administrators a way to track what has been built and reduce duplication.

Governance is becoming a key battleground for cloud providers as customers move from AI experimentation to operational use. Buyers are asking not only whether an agent can complete a task, but also whether it can be audited, constrained and placed within an organisation's broader risk framework.

Monitoring agents

The release also brings together tools for observing what agents do and judging how well they do it. Agent Observability provides tracing and dashboards around reasoning, tool use and execution, while Agent Evaluation is designed to monitor production performance and detect drift over time.

That pairing highlights another issue gaining prominence in enterprise AI: the need for ongoing measurement after deployment. In practice, companies often find that a model or agent behaves differently in production than it did in testing, especially when it interacts with live data, changing prompts or connected tools.

Google Cloud cited several customers as users or evaluators of parts of the platform, including AT&T, Best Buy, Commerzbank, Liberty Global and WellSky. Their comments underline commercial interest in applying memory, identity and governance controls to agent systems used across customer service, platform engineering and internal compliance processes.

AT&T described its use of Agent Memory Bank in sales-related applications. "At AT&T, as we are leveraging Agent Memory Bank for long-term memory, our autonomous & intelligent AI sales agents in the app channel can resume conversations after a gap by synthesizing key facts from prior customer interactions, effectively moving from guessing to remembering. As we extend this capability to IVR [Interactive Voice Response], we're building toward a seamless cross-channel sales journey where customers can continue conversations across app, voice and web experiences without losing context or having to repeat themselves," said Jeff Dixon, AVP Digital Product Management & Development, AT&T.

Best Buy focused on security and accountability for non-human actors inside corporate systems. "At Best Buy, we see as more organizations adopt AI agents, agent identity is becoming just as important as human identity. In the past, we've struggled with orphaned service accounts, unclear ownership, and permissions that kept growing over time. Agent Identity helps bring accountability and governance to autonomous systems by making it clear who an agent is, what it can access, and who is responsible for it. From a security standpoint, applying least-privilege access to agents reduces risk while giving organizations the confidence to scale AI safely," said Kishor Patil, Senior Manager, Cloud Platform Engineering, Best Buy.

Commerzbank highlighted the governance side of the platform. "At Commerzbank AG, we are building a secure foundation for responsible Agentic AI on Google Cloud. To bring this vision to scale, we are actively evaluating Google Cloud's new Agent Registry and Agent Gateway Services. These services are key to our governance strategy, offering vital controls for agent discoverability, policy enforcement, and access management. Furthermore, they provide the deep observability and auditability essential for us to scale our AI platforms in a compliant and trustworthy manner," said Seenuvasan Devasenan, Cluster Architect / AI Transformation Office, Strategisches Programm AI, AI Platforms & Services, Commerzbank AG.

The launch extends Google Cloud's effort to define a fuller software stack for enterprise AI agents, as large technology companies try to move beyond model access into the infrastructure, control and management layers around deployment. "At WellSky, responsible AI scaling means staying ahead of governance. As we expand our Gen AI capabilities across health and community care platforms, our platform engineering team established a proactive framework to catalog, version and lifecycle-manage AI agents in our ecosystem. Partnering with Google Cloud, we've implemented a centralized Agent Registry that enforces compliance policies and ensures only fully vetted agents reach production, while remaining architected for flexibility as our technology evolves. The result is the foundational visibility our teams need to accelerate AI innovation without compromising the security and governance standards our healthcare clients depend on," said Joel Dolisy, Chief Technology Officer, WellSky.