Generative AI Chatbots: IoT Relevance

April 12, 2024

Last year, we all watched as Generative AI (GenAI) took the world by storm writing outreach emails, summarizing encyclopedias of information, and inserting rainbows into photos. Things got so popular, that a new job description appeared on resumes called “prompt engineering.” Prompt engineering was the delicate balance of asking a GenAI tool a question in multiple nuanced ways until the GenAI spit back just the right combination of words. Like almost everyone else, I was amazed by some of the things we were starting to see.  

We started thinking, “How does this apply to me and IoT?” 

Support

Almost everything we initially saw from GenAI came in a chatbot interface, quickly supporting user questions with well-written summaries. The immediate analog for many of us was that what if GenAI could answer questions about how to solve problems using a large set of documentation and data from IoT? Imagine your plumber walked into your house and asked, “What’s wrong with the kitchen faucet?” and the response was, “It looks like there is sediment buildup in the downstairs water heater,” followed by human-friendly instructions on how to flush that particular brand and heater model. It’s a pretty cool way to reduce the diagnosis and maintenance process of the issue. While much of that solution could work without GenAI providing the answer, in a way that would work for any kitchen sink on any house with any water heater.

Display

GenAI helps us look at our data faster. GenAI creates text in formats we like to read, with mountains of historical information about how we want to look at our data and can produce meaningful and accurate interfaces quickly. ClearBlade just announced its first version of GenAI in our Intelligent Assets product, where we are introducing this GenAI chatbot capability. Now, you can ask the GenAI assistant to “please create five trucks to monitor,” and the associated data model will be built with built-in suggestions for common attributes of interest. Future iterations are already showing success at building that dashboard your customer service center is requesting and deploying that rule to alert the field operator when things have gone wrong.

Integration

The hardest part of IoT has been getting things installed and connected. A prolific number of protocols with settings for converting millivolts, tags, or registers into something a human can read makes each environment unique. On the other side, every company has picked a different ETL tool, database, or customer relations management tool, meaning that data, once ingested by IoT, must be converted repeatedly. GenAI, with its ability to rapidly make sense of many different integrations, has the opportunity to significantly reduce the effort to build all these protocol and middleware connectors. Imagine simply handing your GenAI assistant the manual to your new pump and letting it create the code for pulling the appropriate Modbus registers and mapping them directly into your Intelligent Asset instance.  

Summary

With the ClearBlade GenAI Assistant launch, we see what’s possible with GenAI chatbots and GenAI more broadly in IoT. We see a huge opportunity to reduce the time it takes to configure new environments while providing significantly more powerful tools to support our systems’ end operators.

A huge shout-out goes to the AI, Intelligent Assets, and Design teams for collaboratively integrating this powerful capability into an understandable, intuitive framework that will allow us to continue to innovate for many years ahead.

Watch a demo of the GenAI Assistant for Intelligent Assets!

Reach out to the ClearBlade team

 

Schedule A 30 Minute Demo