16 sources. Less than a second. $0. Welcome to the future.

Yesterday, I tried xAI’s Grok chatbot for the first time. I was amazed how well it performed — but can it beat ChatGPT?

I decided to run a head-to-head test to find out.

I will ask the same 3 questions to Grok and ChatGPT*. Let’s see who wins!

Prompt 1: Info on Databricks

Databricks just announced a massive, $10 billion funding round at a $62 billion valuation. When I saw that headline, I wondered, “How long did it take them to get this big?”

Let’s ask Grok…

The question is straightforward and simple. But the quality of Grok’s response stunned me.

Even for this simple answer, Grok cited 15 websites and an X post. I can easily browse each of them, confirming that Grok is correct.

Now, let’s try ChatGPT…

ChatGPT is correct as well. But it doesn’t cite a single source.

If I wanted to make sure ChatGPT isn’t mistaken, I’d still have to Google it. So, what use is ChatGPT?

Grok wins this one easily.

Prompt 2: My (Very Minor) Acting Career

One of my hobbies is being an extra on TV shows and movies.

Yesterday night, I was chatting with a friend who loves Law and Order: SVU. She was surprised when I told her I was actually in it as an extra!

She asked which episode, so I wanted to pull the exact one and link her to it. I gave Grok the title as best as I could remember it.

Let’s see what it says…

Wow! It grabbed the exact episode, gave us a synopsis, and cited a total of 21 sources. These sources are highly relevant — one of them is the IMDb page for the episode.

I sent her the link to the IMDb page. Now, she can watch the episode at her leisure, and so can you (I’m in the opening scene on the subway).

Let’s try ChatGPT…

ChatGPT’s response stunned me.

It found the episode and linked to the IMDb and a Law & Order Wiki. That’s about the same as Grok. But what came next was wild…

ChatGPT included a YouTube video showing a preview of the episode! I’ve never seen ChatGPT respond to a prompt with a video like that. Amazing.

As good as Grok’s response was, this round goes to ChatGPT.

Prompt 3: Multipart Question on Trains

This one was a toughy. I thought I just might stump Grok.

Yesterday, I wrote about train systems in the US and Japan on this blog. As I was writing that post, I needed some info on train speeds in the two countries.

Let’s ask Grok…

Grok nailed it! Despite the complex, multipart question, it gave us a great answer.

The speeds and distances it cites for trains in the US and Japan are quite accurate. The sourcing is elaborate, pulling 19 citations.

We’re tied between ChatGPT and Grok right now, one to one. Can ChatGPT take the day?

Let’s find out…

I asked ChatGPT to find a train route in America that’s about the same length as the Tokyo-Osaka route.

Grok gave us Philly to Boston, which is almost exactly the same distance as Tokyo-Osaka. ChatGPT is giving us DC to Boston, which is about 50 miles longer than the Japanese route.

Not a bad response, but not as good as Grok. What’s more, ChatGPT didn’t cite a single source.

Once again, I have no idea if this information is right. Since I don’t want to make up facts on the blog, I’d have to run down every detail on Google…or perhaps Grok.

Grok takes this one, easily.

Wrap-Up

Grok beat ChatGPT convincingly in this competition, taking 2/3 head-to-head comparisons.

Grok did a wonderful job with sourcing, giving accurate citations for everything it told us. It also ran faster than ChatGPT, which is important to anyone who does many queries per day like I do.

Up until yesterday, I’d never tried Grok. “How good can it be? Besides, I already have ChatGPT,” I figured.

Oh Francis, how little you know.

Grok is so good that this morning, I changed my homepage from ChatGPT to Grok. Until OpenAI releases the next big thing, Elon’s chatbot rules the roost.

What do you think of Grok?

*For ChatGPT, I used the 4o model. There are more advanced models like o1-preview, but I’ve actually gotten better results in the past with 4o. I haven’t tried o1-pro yet. O1-pro cannot access the internet, so it would likely have performed worse than 4o for these questions.

More on tech:

ChatGPT Search vs. Google: Which Should You Use?

Llama 3.1 vs. ChatGPT: Battle Royale

Meet My Latest Investment: Recall

Save Money on Stuff I Use:

Fundrise

This platform lets me diversify my real estate investments so I’m not too exposed to any one market. I’ve invested since 2018 with great returns.

More on Fundrise in this post.

If you decide to invest in Fundrise, you can use this link to get $100 in free bonus shares!

Misfits Market

I’ve used Misfits for years, and it never disappoints! Every fruit and vegetable is organic, super fresh, and packed with flavor!

I wrote a detailed review of Misfits here.

Use this link to sign up and you’ll save $15 on your first order.