Improve the factuality of your generative AI apps by grounding responses in your data

Foundation models offer incredible answer-generation capabilities, but they can sometimes produce inaccurate or irrelevant outputs. This session explores how grounding AI responses in data can reduce hallucinations and improve the reliability of your applications. We'll delve into the differentiated grounding capabilities offered by Vertex AI, including out-of-the-box grounding capabilities, grounding on public datasets, and grounding in Google Search that can improve the accuracy, relevance, and trustworthiness of generative AI applications.
Speakers: Tom Beyer , Lewis Liu, Spencer Chan
Watch more:
All sessions from Google Cloud Next → goo.gle/next24
#GoogleCloudNext
Event: Google Cloud Next 2024

[@kalubachikonde5360]

Grounding with google search or vertex AI search is great but it is not conversational. Responses are short and straight to the point. How can you allow the LLM to maintain a friendly and conversational tone while using grounding?