When it comes to providing real-time information, ChatGPT still has some limitations. As I write this:

• ChatGPT 3.5 is trained until September 2021 and can’t search the web without the Web Pilot Plugin (plugins… remember those from so long ago?)

• ChatGPT 4 is trained up to April 2023, can search the web, and is semi-reliable but unless you’ve got a clever GPT or custom instructions running is maybe not as detailed as you might want in the real estate profession.

If you are relying on more accurate and thorough research or real-time information, you might want to try Perplexity AI.

I’ve been kicking the tires on it over the past few weeks, and I have to tell you, I kinda like it. 

Tool of the Day: Perplexity AI

Perplexity feels like it’s somewhere in between ChatGPT and what Google Search should be.

It leverages both GPT-3 and 4 as well as a wide variety of other sources on the web - and if you have a paid Perplexity plan, you can also turn on Copilot, which will help you with an even more thorough understanding/research of the topic at hand.

This is in contrast to, say, ChatGPT, which is great at natural language conversation and can create content in a flash - but may sometimes sacrifice accuracy for what sounds good (as we all know!)

Let’s look at the differences in a practical example

If I want to know what the current days on the market are in Broadbeach Waters, here is a simple query on ChatGPT 4:

In contrast, in Perplexity, it will look like this:

As you can see, I have a much more detailed answer, along with various sources, so I can double-check the information given.

Let’s now look at the result with Copilot turned on, and I get even more detailed info and an even wider variety of sources.

In this example, I asked the question, and then Copilot asked me whether I wanted to know about houses or units.

I answered ‘houses’, and here’s what it gave me:

I also like the intuitive way Perplexity asks follow-up questions - for example, “How has the stock on the market for houses in Broadbeach waters changed in the past year?” and I get some pretty detailed answers from another 8 sources.

Contrasting all three responses, it feels like we’ve entered a new zone where it's no longer a one-size-fits-all when performing different tasks with AI.

Clearly, each model has its own strengths and weaknesses.

ChatGPT, Bard, and Claude are still clear winners for me for creating and summarising content.

But when it comes to research, I think Perplexity is something I’ll definitely be adding to my toolbox.

Happy Hunting 🚀