Google is launching Agentspace. That’s a solution that securely combines Gemini with corporate data, can search, and take actions.
Google introduces Agentspace. Agentspace is a custom enterprise solution in which Google combines its Gemini LLM with its proprietary search technology, enterprise data, and the ability to take actions.
In fact, this is a slightly expanded version of Gemini, but the evolution of AI is driven not only by actual progress but also by buzzwords. This period, every update for an AI capability seemingly has to have “agent” in the name, and Google doesn’t want to be left behind.
Reasoning through prompts and business data
Agentspace still works based on prompts in natural language, but is capable of further reasoning on the prompts. The AI solution can search corporate data for relevant information wherever it is located. This allows employees to bring all relevant information to the forefront with a single prompt, without having to manually search through different systems. Agentspace works with structured and unstructured data, even when documents are available in different languages.
Google indicates that at launch, Agentspace can be paired with Confluence, Jora, Microsoft SharePoint, ServiceNow and its own Google Drive, among others. Agentspace and Gemini 2.0 are also included in the new NotebookLM Plus, which is available in early access.
Agents
Google Agentspace also integrates with systems and APIs to take actions at a user’s request. Low-code solutions will allow employees to automate workflows, and then deploy those automations through the Agentspace interface. Google shows a demo in which Agentspace summarizes tickets from Jira and sends an email via Outlook based on them, and another example in which the tool helps create an output report.
Google makes the case that Agentspace is relevant to all profiles within an enterprise. The solution, according to the company, can:
- Software engineers help (but so could Gemini and other AI tools);
- Assist marketers (as before, but now with faster access to corporate data);
- Analysts help see trends (again by linking business data);
- Assist HR specialists (with automations, for onboarding, for example).
Google insists that Agentspace is secure, with auditable access to corporate data and guarantees of compliance. Agentspace is available in early access.
Agentspace, Agentforce and Copilot
The above may not sound all that new. For starters, the name (presumably not coincidentally) is very reminiscent of Salesforce’s Agentforce. Agentforce, announced at Dreamforce, fulfills a similar role but in the Salesforce ecosystem. Further noteworthy is that functionality with which Google unpacks is reminiscent of Microsoft’s Copilot integrations. Copilot combined with Power Apps, for example, already brings AI and automation together.Only Microsoft named that functionality before the word “agent” became hype.
One thing is clear: Whereas LLM-based AI was supposed to assist passively just last year, the trend has since shifted to activation. AI must be present in multiple (or preferably all) systems, in a secure manner, and be able to handle complex prompts there. Moreover, according to Google, Salesforce and Microsoft, users should be able to at least initiate simple tasks through the LL interface. Agentspace fits perfectly within that trend.