Google Agentspace Enterprise brings AI agents and AI-powered search to enterprises.
Information discovery across the enterprise: Agentspace Enterprise gives employees a single, company-branded multimodal search agent that acts as a central source of enterprise truth for your entire organization. Building on the best of Google's search capabilities, Agentspace Enterprise can provide conversational assistance, answer complex questions, make proactive suggestions and take actions based on your company's unique information.
Agentspace Enterprise works across both unstructured data – such as documents and emails – and structured data such as tables. Google also built translation in so that you can understand information, even if it originates in another language. With prebuilt connectors for the most commonly used third-party applications, like Confluence, Google Drive, Jira, Microsoft Sharepoint, ServiceNow, and more, your employees can access and query relevant data sources and make better decisions.
Expert agents to automate your business functions: Agentspace Enterprise Plus offers expert agents to automate your business functions. Agentspace Enterprise Plus is the launch point for custom AI agents that apply generative AI contextually. Enterprises can empower their employees in marketing, finance, legal, engineering, and more to conduct better research, draft content quickly, and automate repetitive tasks, including multi-step workflows. With Agentspace Enterprise Plus, enterprises can scale AI by hosting these agents in an intuitive interface where employees can discover and access all expert agents in one place.
Why use Agentspace Enterprise?
Agentspace Enterprise is an intranet search, generative answer, and agent solution. In most enterprises, knowledge discovery tools don't offer personalization, answer generation, user context, or deep retrieval. Users lose time searching across a variety of systems. Agentspace Enterprise can empower employees to find the right information at the right time by connecting content across an organization and generate grounded, personalized answers from it. With Agentspace Enterprise, you can:
- Ingest large volumes of enterprise data from multiple formats and platforms into a single app
- Sync and search across data from SaaS systems such as Salesforce, Jira, Confluence, and more
- Scalably deliver access-controlled search results and generative answers to thousands of employees
- Streamline agentic workflows by working across your enterprise data, eliminating the need to switch between multiple AI assistance and agent applications by integrating them into one workflow.
Key pillars of Agentspace Enterprise
Agentspace Enterprise builds its features on:
- Built-in trust: Prioritize security, explainability, and governance to ensure reliable and transparent results. Authenticating experiences with SSO with a built-in permissions model and support for user-level access controls.
- Google intelligence: Learn from user behavior and content patterns to deliver results that are not only relevant but anticipate the unasked questions. This means going beyond keyword matching to deep semantic intelligence, powered by Google's planet-scale language understanding, knowledge graphs, and large language models.
- Universal connectivity: Connect to various enterprise systems with on-demand and automated refresh, syncing to break down barriers between cloud platforms, legacy systems, and file shares.
- Enterprise-level customization and user-level personalization: With advanced controls for enterprise with search, recommendations, LLMs, and knowledge graphs, provide a personalized search experience for each user that factors in their role, permissions, and search history.
- Real-time feedback and adaptation: Continually improve results based on real-time feedback with large-scale, machine-learning-based systems.
- Blended RAG: Customizable RAG over a blended data source. Power generative AI use cases that are grounded on enterprise data spread across multiple formats and platforms.
- Scalability: Scale across geographies, across languages, and across peak volume requirements.
Start using Agentspace Enterprise
To start using Agentspace Enterprise:
- Set up data source access control: Google uses your identity provider to identify the end user performing a search and to determine if they have access to the documents that are returned as results. You must have your organization's identity provider configured in Google Cloud. Then, connect your identity provider in the console and grant search permissions to your users.
- Import or connect data: Create data stores or data connectors for
each enterprise data source.
- Create a data store: To use data from Google sources such as BigQuery, Cloud Storage, and Google Workspace, create a data store and import your data into it.
- Connect a third-party data source: To use data from third-party sources, for example Jira Cloud, Salesforce, and ServiceNow, create a connector that syncs your data.
- Create a Agentspace Enterprise app: When you create a Agentspace Enterprise app, you can connect it to the data stores and data connectors that you've created. Your app can then search across all of your enterprise data.
- Create and use assistants: Create and configure assistant for your app to offer generative answers and conversation that's grounded on your enterprise data, and optionally, grounded on public website data.
- Turn on the web app: The Agentspace Enterprise web app provides you an out-of-the-box user interface for your search app with data source and recency filters, knowledge cards, and generative assistant features.
- Preview search results: You can preview how search results appear in your app using the console, or get search results programmatically.
Configure Agentspace Enterprise
Agentspace Enterprise provides many options for configuring your search app and results:
- Preview search results. To get search results programmatically with additional parameters, see the code examples in Get search results.
- Assistant: You can create an assistant for your app to offer generative answers and conversation that's grounded on your enterprise data. See Create and use assistants.
- Serving controls: Control when search results are boosted, buried, filtered, or redirected, or whether certain queries are associated with other queries as synonyms. See Configure serving controls.
- Answers and follow-ups: Agentspace Enterprise supports generative answers and follow-ups with the Answer API. See Get answers and follow-ups.
- Field settings: For first-party structured data or for unstructured data with metadata, update field settings to refine how metadata is used for search. See Configure fields for search.
- Search analytics: View search analytics from your app in the Agent Builder console. See View analytics.
Configure and connect data sources
You can configure data source access control and ingest or connect data from Google and third-party data sources:
- Identity and permissions. The administrator for your organization's identity provider should set up your identity provider with Google to apply access control to your results, and review the permissions to grant to third-party data sources if you plan to use them. Go to Identity and permissions.
- Third-party connectors. The administrator for the third-party applications (such as Jira or Salesforce) that your organization uses can review and obtain the credentials needed for Google to access third-party data. Then, ingest data from third-party applications to use with Agentspace Enterprise. Go to Connect a third-party data source.
- Google data ingestion. Ingest data from first-party applications such as Cloud Storage and Google Workspace. Go to Create a first-party data store.