Google has just introduced Agentspace. According to the company, it will let users perform complex tasks with just a single prompt.
For now, Agentspace is in early access. Those interested can sign up here. The package that awaits them is significant, at least in what it promises. For example, Google says that complex information can be retrieved via single prompts from an employee. NotebookLM is the key to success here. With the newly unveiled NotebookLM Plus, employees can upload information to have it synthesized, possibly even in the form of a “podcast-style audio summary.”
Central agent
Agentspace provides employees with a multimodal search agent. It is said to not only handle structured data well, but can also provide the right documents, emails and more where it is relevant. Pre-baked connectors with Confluence, Google Drive, Jira, SharePoint and ServiceNow, among others, ensure that the search function instantly brings together a broad spectrum of tools.
Just looking up data is one thing. But the power of Agentspace is that it can reduce the overall workload for employees. That’s why Google characterizes the new offering as the “launching point” for custom AI agents that apply generative AI based on context. Examples include repetitive tasks for marketing teams or legal matters, allowing human employees to focus on the trickier stuff. It’s the same story we’ve heard before when it comes to agents, although Google intends to be the centre of its deployment due to its wide scope.
Already a lot of enthusiasm
Some customers are already raving about Agentspace. For example, Chief Digital Officer at Nokia Alan Triggs sees the broad support for search as a key benefit. “We’re particularly excited by Google Agentspace’s ability to blend various data sources quotes and deliver personalized, contextually relevant answers. By unifying our knowledge resources, providing AI powered assistance and automating workflows, we strive towards reduced time spent searching for information, faster decision-making, and improved collaboration and productivity.”
Fabio Jabut, Head of Data and AI at Spain’s Banco BV, is also enthusiastic. “We are particularly excited to see how our early users are leveraging Google Agentspace for faster and more comprehensive analysis and engaging with content in new multimodal ways that reflect our vision of work as more relaxed and collaborative.”
‘Agent of agents’?
This move by Google is arriving early enough in the adoption of AI agents to contend for a future standard. While Microsoft is still dawdling with agents that aren’t really agents, Google is working on integrations that should make the rollout of decisive AI assistants smooth sailing in the end. Multi-platform support makes it an attractive starting point for organizations as well. Still, we wonder if parties like Salesforce, which itself wants to be the centre for agents on their platform, will cooperate with this kind of setup. It seems obvious to want to create an agent management platform within the suite you’ve already adopted, such as Microsoft 365 or Google Workspace. But in the land grab of AI solutions, such centralization leads to fewer opportunities for others, who would eventually have to integrate with the bigger players.
It remains to be seen whether Agentspace is the blueprint for running AI agents. For now, it is a very early stage for GenAI at the enterprise level, where organizations may not have even figured out what they intend to do with the technology to transform their way of working.
Also read: AI agents need Process Intelligence