For productive AI to reach its enormous potential, it must be widely accessible and easy to integrate into a range of services. Vertex AI Search and Conversation allows developers with minimal machine learning expertise to create and deploy smart apps in as little as a few hours.

Introduced in preview earlier this year as Enterprise Search on Generative AI App Builder and Conversational AI on Generative AI App Builder respectively, Vertex AI Search and Conversation offers a simple management layer to combine enterprise data with generative core models as well as dialogical AI and information gathering technologies.

Instead of spending months building AI applications, enterprise developers can quickly retrieve data, add customization, and create a search engine or chatbot that can interact with customers with a few clicks and answer questions based on the accuracy of specific structured and unstructured data sources, as well as corporate websites. The ability to quickly prototype productive applications allows businesses to track a variety of use cases, from food ordering to banking assistance and customer service.

In addition to general use, today we are introducing new features so that developers can not only allow users to find important information through natural language, but also create even more engaging applications that can perform actions on their behalf. These new features include:

• Versatile search; providing clear summaries for search results and chat conversations speech and call summarization and tools that allow developers to pre-program prompts and responses for specific queries in natural language just as they instruct a human.
• Capable of receiving information in real time and acting on behalf of users in third-party applications such as Google and Datastax, MongoDB and Redis Vertex AI extensions and Vertex AI data connectors, which help retrieve data from enterprise and third-party applications such as Salesforce, Confluence, and JIRA, connect manufacturer applications to widely used enterprise systems.
• Organizations can flexibly decide whether they want the data to be supported by the training data of the underlying model. They can also use useful features such as citations to increase user confidence in the quality of the results returned.

Let's take a closer look at the capabilities of Vertex AI Search and Conversation.

‍

Build personalized, engaging productive apps with Vertex AI

Vertex AI Search, Allows organizations to install Google Search-quality, multi-mode search applications supported by basic models. It also offers the ability to base outputs solely on enterprise data and use enterprise data to complete initial training of the basic model. It supports features such as citations, relevance scores, and summarization to encourage enterprise access controls and trust in results to ensure information is only available to eligible users.

Organizations with more complex use cases can combine LLM layers with vector search to power a wide range of productive AI applications, such as semantic search, personalized recommendations, chat, multimodal search, and more. Vector search gives organizations access to an easy-to-use vector similarity search solution, the same technology that Google uses to power major services like Google Search and YouTube at scale.

‍

‍

Vertex AI Conversation, facilitates the creation of human-like chatbots and voice bots with a natural sound, supported by basic models with both voice and text support. With this, developers can create a chatbot based on a website or collection of documents in just a few clicks. For further customization, Vertex AI allows developers to combine decisive workflows with productive outputs, combining rules-based processes with dynamic AI to create engaging yet reliable applications; including processing capabilities, users can ask AI representatives to make appointments or make purchases, for example. Organizations can set up conversations with various data from websites, documents, FAQs, emails, and representative conversation histories, and generate interaction summaries, excerpts, and other data to facilitate transfers between AI applications and human representatives.

‍

With the power of Google's core models and user-friendly developer tools, organizations can:

• Personalized experiences for customers and employees creating tasks that used to take hours, turning them into quick searches or conversational explorations. From uncovering the right information in the right context to executing actions, creating images, jotting down quotes, and generating suggestions, apps can respond naturally to user input, giving organizations control over tone, speech flows, data access, and more.
• Data collection, creation of associations or eliminates the need to manage indices and speech decision trees. Instead, developers can take advantage of a simple editing interface to quickly build applications with little or no coding and no prior machine learning experience.
• Basics, helps maintain control over application actions and data. Developers can increase the relevance of outputs and reduce illusions by basing outputs on enterprise data. They can also control what data apps can access through connectors and what actions apps can perform on behalf of a user. The data is stored in an organization's own instance of Google Cloud, and Google does not access this data or use it to train our models.

‍

With Vertex AI Search and Conversation, organizations can achieve significantly faster deployment time and value acquisition time.

Omar Omran, Head of Digital at Six Flags, highlighted the transformative potential of collaborating with Google Cloud and using Vertex AI Conversation. "With Google Cloud's technology, we are committed to not only improving parking operations, but also creating unique, personalized guest experiences. By bringing a higher level of agility and responsiveness to our operations, we will redefine the way we serve our guests and set new benchmarks in the amusement park industry.”

According to Harsh Kumar Sarohi, Senior Vice President of Technology at Tradeindia.com, one of India's largest B2B portals, “Using Vertex AI Search in Google Cloud, we were able to reduce our search drop rate by 50% and increase existing user engagement by 6%. We also leverage Vertex AI Search's analytics capabilities to understand gaps in our existing portfolio using metrics such as best unqueried search results.”

C1, a leading provider of communication center solution technologies, uses Vertex AI Conversation to support its employees as well as its customers. The agent support bot C1 Auto Pilot can help agents improve the customer experience by providing automated summarization with real-time customer engagement recommendations, cross-sell/top sales opportunities, sentiment analysis, and call insights. According to Mark Langanki, C1 CTO, “The overall goal of the C1 Auto Pilot solution is to help agents improve the customer experience by reducing the time they spend on administrative tasks such as note-taking and research, and to provide the information they need to have a more efficient and engaging conversation with the customer. We have seen reduced call processing times by providing agents with customer-specific value-added information. We expect continued improvement in both agent experience and customer satisfaction.”

‍

Search and conversation use cases provide a clear opportunity for organizations to quickly gain experience and leverage productive AI technologies.