Google Cloud has recently introduced a powerful new feature: Grounding with Google Search, now available in both the Google AI Studio and the Gemini API. This feature is designed to improve the accuracy and freshness of responses generated by the Gemini AI models. By integrating Google Search into the AI’s response process, developers can ensure that their applications provide users with more up-to-date and relevant information.
The primary benefit of this integration is the enhanced accuracy of the generated responses. Not only does the AI model deliver more precise answers, but it also provides users with links to the original sources that ground the responses. This feature increases transparency, allowing users to trace the information back to its origin. Additionally, Google Search offers search suggestions that guide users to related results, further enriching the experience by providing broader context.
Grounding with Google Search is available for all generally accessible versions of the Gemini 1.5 models. Users can easily access it in the Google AI Studio under the “Tools” section or activate it within the Gemini API by enabling the google_search_retrieval tool. While users can experiment with the grounding feature for free in Google AI Studio, developers seeking to use it through the Gemini API will need to subscribe to the paid tier, which costs $35 per 1,000 grounded queries.
The advantages of grounding extend beyond just factual accuracy. By accessing real-time information from Google Search, the AI models become more adaptable to various scenarios, ensuring their relevance in an ever-changing world. This ability to retrieve fresh, accurate data allows AI applications to stay aligned with current trends and developments. Furthermore, the integration of supporting links directly into the responses enhances the overall transparency and credibility of AI-powered tools, building greater trust among users.