Knowi is always looking to bring innovative features to its users, with the latest one being its integration with OpenAI. This has unlocked doors for:
I. Creating a dataset with OpenAI as a datasource and developing visualizations on top of it, and
II. Conducting data exploration and generating queries using plain English statements.
TL;DR
- Knowi integrates with OpenAI in two ways: as a direct data source for generating AI-powered datasets, and as an AI query generator for writing queries in plain English.
- You can connect OpenAI as a datasource in Knowi, send prompts, and build dashboards on the AI-generated responses.
- The AI Query Generator lets non-technical users describe what they want in natural language, and Knowi translates it into executable queries.
- Knowi also offers Document AI for chatting with uploaded files and NLQ across all connected data sources.
- All AI features work with Knowi’s Private AI deployment, keeping data on-prem with no external API calls.
Table of Contents
What is OpenAI?
OpenAI is an AI research company that develops large language models and AI tools, including GPT-4 and the ChatGPT platform.
The mission is to ensure that artificial general intelligence (AGI),by which the company means highly autonomous systems that outperform humans at most economically valuable work,benefits all of humanity.
The organization is dedicated to conducting fundamental research in AI, building cutting-edge AI technologies, and advocating for safe and ethical AI practices.
Knowi’s OpenAI Datasource Integration
Knowi provides the option to create a dataset with OpenAI as a datasource and develop visualizations on top of it.
Before you can create your dashboard with OpenAI, you need to add OpenAI as a data source and query against it.
For more information, you can follow OpenAI datasource documentation.
Knowi’s AI Query Generator
The AI query generator from Knowi automatically generates queries from plain English statements, searches the connected databases, and retrieves the data requested.
The goal for this integration involves simplifying and speeding up the search process by automatically generating relevant and specific queries and improving the probability of finding the requested information.
The following example displays a MongoDB query being generated within Knowi using plain English statements. In the illustration below, you can observe how the MongoDB database is being queried to-
“Return spam rate, spam, and sent from the sendingActivity collection, grouped by customer. Spam rate is calculated as (spam/sent)*1000000”
The AI Query generator has been integrated with multiple datasources including, but not limited to, Snowflake, ElasticSearch, MySQL, Couchbase, PostgreSQL, and more.
AI Integration Options for Analytics Platforms
| Feature | OpenAI API (standalone) | ChatGPT Enterprise | Knowi with OpenAI |
| Data access | Requires custom code to connect to databases | File uploads only, no live database queries | Connects to 30+ databases natively, sends context to OpenAI |
| Visualization | None (text/code output only) | Basic charts from uploaded data | Full dashboard builder with 20+ chart types |
| Natural language queries | Must build custom NL-to-SQL pipeline | Asks questions about uploaded files | NLQ translates plain English to native database queries |
| Data privacy | Data sent to OpenAI servers | Data sent to OpenAI servers (enterprise tier) | Private AI option: run models on-prem, no data leaves your environment |
| Embedding | Build custom UI from scratch | Not embeddable | White-label embedded dashboards with AI built in |
| Multi-source blending | Manual via code | Not supported | Cross-source joins across SQL, NoSQL, and APIs |
Conclusion
Knowi implemented the OpenAI integration to help data engineers, data analysts, data scientists, or anyone responsible for managing the data to make querying easier while increasing overall productivity. Further more, OpenAI as a datasource within Knowi allows you to enrich the existing data with OpenAI, whether it be in your query workflow or dashboards.
Want to see how Knowi integrates AI into your analytics workflow? Schedule a demo to explore Private AI, NLQ, and OpenAI integration firsthand.
Frequently Asked Questions
How does Knowi integrate with OpenAI?
Knowi integrates with OpenAI in two ways: as a direct data source where you can send prompts and build dashboards on AI-generated responses, and through an AI query generator that translates plain English into executable database queries.
Can I use OpenAI with Knowi without sending data to external servers?
Yes. Knowi’s Private AI deployment runs models on-prem so no data leaves your environment, making it suitable for regulated industries like healthcare and finance.
What databases work with Knowi’s AI query generator?
Knowi’s AI query generator works with 30+ databases including MongoDB, Snowflake, Elasticsearch, MySQL, PostgreSQL, Couchbase, and more.