LinkedIn
Twitter
Google's Vertex AI platform is a powerhouse for building cutting-edge AI solutions, and one of its standout products is Vertex AI Search and Conversation. But what’s the difference? and how do you know which one to use for your business needs? Let’s break it down.
Sample Demo links for the entire procedure: Search: Search_Demo_Full.mp4, Chat: Chat_4k_Blog_Mask.mp4
Youtube Shorts link: https://www.youtube.com/shorts/C5WnUur_N1I
Before getting into Vertex AI Search vs. Vertex AI Conversation, let’s get into Search vs. Chat. The easy way to understand the difference is this:
Search, like a librarian, gives you a list of citations that might contain the answer, possibly with the summarized response to the specific question.
- One and done: You ask, it answers, and then you’re on your own if there is no ‘follow-up’ enabled.
- Great for: Finding information and related websites, documents, or specific citations.
Conversation, like a friend, initiates a back-and-forth chat where someone understands what you mean over time, i.e., context.
- Flexibility: The conversation can change and proceed in different directions as you chat based on intents, routes, and flows.
- Great for: Getting help with tasks, having open-ended discussions, exploring ideas, or just having a fun conversation.
When to Choose What: Vertex AI Search vs Vertex AI Conversation?
Now, think 'Search vs. Chatbot' from the perspective of Vertex AI. While they both rely on powerful Google AI technologies, Vertex AI Search and Vertex AI Conversation serve slightly different purposes.
- Vertex AI Search is like a supercharged search engine designed specifically for enterprise data. Need to find the relevant context and citations buried in a mountain of files? Vertex AI Search has you covered, as it excels at providing search and discovery across data by combining search and reasoning to answer complex questions based on the organization's data corpus.
- Vertex AI Conversation is your AI-powered chat agent or virtual assistant that facilitates the creation of human-like conversational experiences. It can understand user questions, provide helpful answers based on enterprise data, and even complete tasks like booking reservations or checking account balances (via Dialogflow CX webhooks, generators, etc.). It enables chatbot developers to zero in more on conversation design and less on particulars of execution and business logic.
Both products are capable of helping your ‘question’ find its ‘answer’ needle super-quickly in the enormous haystack of your documents, websites, structured data, etc. So, search and chat are possible by meaning as well, not only the ‘keywords’.
Use cases
- Just want to quickly kickstart your Vertex AI Search and Conversation journey for exploration? -> Vertex AI Search
- Do you want to provide a conversational interface for users to get help or interact? -> Vertex AI Conversation
- Do you need to make information easily discoverable via a search bar or widget? -> Vertex AI Search
- Are you playing with different types of data at once (website, unstructured, or structured data)? -> Vertex AI Conversation
Note: Every search app has a 1:1 correspondence with a datastore, whereas a single chat app can handle multiple types of datastores.
- Do you have FAQs to address? -> Vertex AI Conversation
- Have a BigQuery datastore and desire flexibility around which fields will be retrievable, indexable, and searchable? -> Vertex AI Search
- For the website datastore, if domain ownership verification is a major blocker? -> Vertex AI Search
Note: The chat app mandates advanced website indexing, which further requires domain verification. More details are here.
- Want to leverage Dialogflow CX features for your app (such as flows, webhooks, generators, multiple GA languages, IVR, and so on)? -> Vertex AI Conversation, hands down, as Dialogflow CX is its built-in component!
- How seamlessly do you want the solution to fit into your existing systems? Both products offer APIs to customize and integrate the search bar and chat UI, respectively -> Vertex AI Search and Conversation
Verdict: Both end up with a 5:5 score. There are numerous other fine-grained and intricate distinctions as well. The detailed analysis of your use case is the primary factor that will reveal those and break the tie 🙂.
Under the hood: How does Vertex AI Search and Conversation work?
Both Vertex AI Search and Vertex AI Conversation leverage Google’s advanced foundation models for language understanding and information processing. However, they use these technologies in different ways.
Vector Search Methodology
Vertex AI Search
Vertex AI Search employs vector-based search. This means it turns your text data into numerical representations, allowing it to find information based on meaning, not just exact keywords. Technically, it is based on the world’s largest and most scalable search engine (Google Search ST-MU), which combines both keyword search and semantic search.
- Output control: Free to customize the output through prompt templates, whether extractive segments or snippets.
- Analytics: Search history per session and comparison.
- Technology: Vector-based search and foundation models (the almighty Google Search engine employs similar technology as well!).
Vertex AI Conversation
Vertex AI Conversation combines foundation models with Dialogflow CX, Google’s powerful conversational AI platform. This gives it the ability to understand the nuances of conversation, follow the flow of a dialogue, and provide relevant responses. Thereby, it simplifies complex conversational flows and graphs via natural language prompts.
- Output control: Free to customize the output through generators and external webhook triggers (if you want to perform extra tasks!).
- Analytics: Conversation history export to BigQuery Table (if enabled) and comparison across sessions.
- Technology: Foundation models, Dialogflow CX for conversation flow, State machines, and intent-based NLU (Natural Language Understanding) Technology’.
Sample Results
Let’s look at a few sample search and chat results related to a sample Wiki PDF ‘Tourism in India’, which are as follows:
Search Widget (Source: Vertex AI Search for Wikipedia PDF Search Widget)
Chat Widget (Source: Vertex AI Conversation for Wikipedia PDF Chatbot)
Vertex AI Search and Conversation for RAG (Retrieval Augmented Generation)
Vertex AI Search and Conversation combine to create powerful Retrieval Augmented Generation (RAG) solutions. Vertex AI Search excels at indexing and retrieving relevant information by making it discoverable from vast data sources - documents, websites, or structured data.
Meanwhile, Vertex AI Conversation acts as the generative component, crafting natural-sounding responses based on the retrieved knowledge to foster natural interactions with your customers and employees. Also, Dialogflow employs Enterprise Search to search for sources based on the user's query.
This synergy allows you to build chatbots or virtual assistants that don’t just parrot back memorized responses but instead, can access, understand, and then articulate information from your collective knowledge base.
Vertex AI Search and Conversation: The bottom line
This blog is not about the battle of two heavyweights, as Vertex AI Search and Vertex AI Conversation complement each other and don’t work in isolation. They are both powerful features for making the most of your company’s enterprise data. By using the power of this combination, the beauty of Vertex AI Search and Conversation as a whole product can be realized. By understanding their differences and potential use cases, you can choose the right tool for your specific needs.
Verdict: Neither Vertex AI Search nor Conversation individually win as a feature of Vertex AI Search and Conversation which is a single-win product and provided out-of-the-box (OOTB) as a fully-managed service.
Sample Demo links for the entire procedure: Search: Search_Demo_Full.mp4, Chat: Chat_4k_Blog_Mask.mp4
Are you ready to start your quest? Embark on your Vertex AI S&C adventure and conquer your data chaos using the reference links:
References and additional resources
- [Blog] Vertex AI Search and Conversation is now generally available
- [Documentation] Vertex AI Search and Conversation release notes
- [Documentation] Data store agents | Search configuration
- [Codelab] Use Vertex AI Search on PDFs (unstructured data) in Cloud Storage from a Cloud Run service
- [Google Cloud Skills Boost] Integrate Search in Applications using Vertex AI Search
Let me know in the comments if you’d like a deeper dive into any specific technology or examples of how they’re used in real-world applications! Should you have any concerns or queries about this post or my implementation, please feel free to connect with me on LinkedIn! Thanks!