AI in 5 Minutes: What It Is, What It Isn't

Jess: Hello, I’m Jess Groff, the VP of Product at Inspect Point. I’m joined here with Pat, co-founder and CEO of Inspect Point. We’re kicking off an exciting series on AI and how you can best leverage it in your business. We’re going to go through everything from what is AI, how it’s being used in the industry, and how you can get started with it.

Pat: AI has had a huge impact on the news cycle, at least. Mixed results in action, but still a ton of promise, and I’m super bullish on what it can do. It reminds me a lot of the internet transforming work, life, and play.

Jess: Totally agree. And I think fire protection has already seen quite a bit of investment. I just spoke with someone at a large national contractor who had the title “Head of AI.” So companies are looking for leverage.

Jess: Let’s jump into what this thing actually is.

Pat: AI doesn’t equal magic. Models learn patterns from data. For a long time, AI was personified as human-like, but it’s really a computing model for processing data. Not too much different than how a brain works, I guess. Around 2017, a new model made it much easier to process larger data sets. And with the popularity of ChatGPT, we saw the rapid rise in adoption. At one point, it was the fastest-growing consumer application in history. Since then, we’ve started to see more platforms and types of AI being developed: LLMs, embedded AI, assistants, agents, and more.

Jess: We’ll get more into the specific tools in our next session, but it’s an important callout that when we talk about AI, it’s not just ChatGPT or one tool. If you’re looking to get value from it, you have to pick the right AI for the job. Old analogy, but it applies here: you wouldn’t use a hammer across a job site. Similarly, you’ll likely need to use different tools for different jobs.

Pat: Similarly, it’s important to test and learn. It’s still early days for the tools, and the more you conduct little tests and learn—keep what works, get rid of the rest—the better you’ll get at using AI.

Jess: We wanted to hit a couple of other key points for understanding and using AI as well.

Pat: Yeah. First, structure your data the right way. There are definitely use cases for one-off prompts, but AI is most useful when it has been trained on your data—how you do business. Sometimes we even see these things called “hallucinations” where it makes stuff up. Because of that, you need to keep humans in charge. Most people will hopefully have seen our tool that builds AI into inspections, but the tech ultimately holds that license and needs to be the one signing off on the job.

Jess: You can get into trouble asking ChatGPT to draft a customer email and sending it without reading it.

Pat: You might end up giving them one hell of a discount. We touched on it briefly, but AI is also most useful when embedded where work already happens. ChatGPT and other purpose-built tools are great, but AI appearing in inspection, scheduling, quoting, and service workflows is where you’re seeing it at scale.

Jess: Finally, it’s important to measure business impact, not model coolness. Is your business actually getting better, or are you just using a cool new tool? We talk about this in sporting clays: “All hat, no cattle.” You can have the nicest gun, but you still have to hit the targets to win. And then choose the right partner. That’s where our pitch comes in. We’re passionate about technology and happy to work with you to learn more about leveraging AI within Inspect Point or in your business broadly.

Pat: Alright, we’re keeping these quick and punchy. Go to inspectpoint.com/ai to chat with us directly. We’ll also be back for Episode 2 to see the toolbox in action. We’ll dig into four real-world use cases for four different types of AI tools.