Google Cloud has added S&P Global Adaptive Retrieval to Gemini Enterprise in Public Preview, giving users a natural-language interface for researching financial information available through their S&P Global accounts. The October 30 release brings company financials, filings and earnings call transcripts into the platform’s connected research environment.
What is Gemini Enterprise?
Gemini Enterprise is Google Cloud’s platform for workplace search, AI assistance and AI agents. It connects information held across an organization’s applications so employees can ask questions, find documents and work with answers grounded in connected sources.
Google provides prebuilt connectors for applications including Microsoft SharePoint, Confluence, Jira and ServiceNow. These connections let employees search across different systems through a shared interface, with results governed by their access permissions.
The platform also supports conversational assistance and custom AI agents. Administrators manage connected sources, security policies and user licenses, while employees use the resulting environment to research topics, analyze material and develop work products such as reports.
The S&P Global connector adds a licensed external information source to that model. For teams already working in Gemini Enterprise, it provides a way to bring financial research into the same environment they use to consult other connected business information.
How Adaptive Retrieval works
According to Google’s connector documentation, a question is routed to the S&P Global data agents that the user’s account is entitled to access. Those agents retrieve relevant information, and Gemini Enterprise generates an answer from the data they return.
The supported material includes structured company financials and documents such as filings and earnings call transcripts. This gives the connection relevance to research that requires both reported figures and the explanations accompanying them.
Access depends on the permissions attached to the S&P Global account. Enabling the connector does not grant access to the provider’s entire information portfolio.
The October announcement specifically concerns the availability of this connection in Gemini Enterprise. It marks a new access point for Adaptive Retrieval, rather than the first introduction of the underlying S&P Global offering.
Where it could fit into corporate research
A potential use is preparing a company briefing ahead of a customer, supplier or financing discussion. An analyst could use authorized financial data and company disclosures to investigate a business’s performance, then examine management commentary in earnings transcripts.
Peer comparisons and preliminary counterparty research are other possible applications. Their usefulness would depend on the datasets available to the account and the information returned for each question. These are potential workflow applications, not customer results reported with the release.
Setup, data handling and preview limits
Google’s setup guide requires an authorized administrator to create a data store, attach it to a Gemini Enterprise app and authorize the S&P Global connection before queries can run. The connector uses Google-managed OAuth, removing the need to create a separate OAuth application or supply a client ID and client secret.
During a search, Gemini Enterprise sends the query directly to the S&P Global Adaptive Retrieval API. Google says the returned results can be combined with results from other connected sources.
This federated search model also has data-handling implications. Google notes that third-party providers may associate queries with a user’s identity. If several federated sources are enabled, a query might be sent to all of them. Information reaching an external system is governed by that provider’s terms and privacy policies.
The connector is supported in the global, US and EU locations. It remains in Public Preview under Google’s pre-general-availability terms, with potentially limited support. Google also documents a deployment constraint: applying a VPC Service Controls perimeter to an existing Adaptive Retrieval data store requires deleting and recreating that store.
Sources
ERP News Editorial Team
The ERPNews Editorial Team covers global developments in ERP (Enterprise Resource Planning), enterprise software, cloud platforms, AI, automation, and digital transformation, providing independent news and editorial analysis for senior business and technology leaders. Our reporting focuses on market signals, strategic shifts, and enterprise impact across the ERP and enterprise technology ecosystem.
For editorial inquiries, please contact:
đź“© [email protected]
