Mountain View, CA – Google Cloud has announced a suite of powerful AI innovations and updates at the Google I/O 2026 conference, aimed at transforming enterprise workflows and developer productivity. The company is making these advancements available to Google Cloud customers through Gemini Enterprise and Google Workspace, signaling a significant push towards an "Agentic Enterprise" blueprint.
Gemini 3.5 Flash and the Agentic Enterprise
Central to these announcements is the introduction of Gemini 3.5 Flash, the latest family of models that combine advanced intelligence with actionable capabilities. This new model is set to roll out in the coming weeks to developers and enterprise customers via the Gemini API and Agent Platform API. Alongside Gemini 3.5 Flash, Google is also introducing Gemini Omni, a new model designed for enhanced world understanding, multimodality, and editing, capable of generating any output from any input, starting with video. These models are part of a broader strategy to equip businesses with tools for the next era of enterprise development.
Enhanced Agent Platform and Code Security
The Gemini Enterprise Agent Platform has received substantial upgrades, including the integration of CodeMender. CodeMender, an AI code security agent originally developed by Google DeepMind, will now autonomously identify vulnerabilities within code, recommend precise fixes, and securely test and apply patches with user approval. This integration aims to automate secure deployment while maintaining developer control. Furthermore, Google is enhancing platform trust and safety for synthetically generated media with a new AI Content Detection API, now rolling out on the Agent Platform. The Managed Agents API on Agent Platform will also allow builders to instantly create custom agents that can reason, call tools, and execute code within secure, Google-hosted remote environments through a single API call.
Workspace Intelligence and Developer Tools
Google is also integrating AI directly into Google Workspace applications, with new features previewed for business customers this summer. These include advancements for Workspace Intelligence, further embedding AI into daily productivity tools. For developers, Google AI Studio now offers native Kotlin support for building Android apps, along with one-click deployment to Cloud Run and Firebase services. Projects can also be seamlessly exported to Google Antigravity, Google's agent-first development platform, which has also seen upgrades with new capabilities for orchestrating and building agents. The company is also highlighting its eighth-generation TPUs and a reimagined Agentic Data Cloud as foundational elements for these AI advancements.
The ongoing development and integration of these AI tools underscore Google Cloud's commitment to empowering businesses and developers to navigate the complexities of the evolving technological landscape. The focus remains on creating more intelligent, automated, and secure solutions for a wide range of applications.