I just tested new Siri vs Old Siri: How smart is Apple Intelligence?
It's time to put the new Siri to the test
Hey, Siri. Are you smart yet?
It's hard to believe that Siri was introduced more than 13 years ago at this point. And during all that time Apple's assistant has been mildly helpful at doing basic things like setting timers and sending messages, but it ultimately got surpassed by Google Assistant (now Gemini), Alexa and now ChatGPT.
So how does Apple respond? By rolling out a new Siri with Apple Intelligence as part of iOS 18.1. The updated Siri has a new design with a glowing animation, and it stays listening so you can continue the conversation. This means that Siri is more contextually aware. In addition, you can now type to Siri and ask all sorts of questions about the iPhone or iOS and get step-by-step instructions.
To be clear, Siri is still a work in progress as other features are still to be added. This includes ChatGPT integration that's coming in iOS 18.2 and the ability to use voice controls with your apps in future updates. But I wanted to know how much smarter Siri is now, so I put an iPhone 15 Pro Max with the new Siri up against an iPhone 11 Pro Max running iOS 17 to see how far we've come.
New Siri vs Old Siri: Find me the nearest steakhouse
I was feeling peckish so I asked both the old and new Siri which steakhouses were nearby. Surprisingly, the old Siri came back with more results in its list, but it was unable to give me the hours for the No. 2 entry on the list; it was also a bit slow. It could get me directions though.
The new Siri was not only faster, but it could get me the hours and directions to the second restaurant without me having to press the side button. This is a benefit of the new Siri continuing to listen, which you can tell by the glow around the screen.
Winner: New Siri
New Siri vs Old Siri: What's the world's tallest building?
I next asked both Siris "what the world's tallest building," followed by "what's the second smallest?" Surprisingly, both the new and old Siri got this right. It's the Burj Khalifa in case you were wondering, followed by the Shanghai Tower — at least until earlier this year when Merdeka 118 opened in Malaysia. That both Siris came up with the same slightly out-of-date answer suggests that they're both accessing a data source that has yet-to-be updated.
Things got interesting, though, when I asked "How about the U.S." The old Siri gave me the tallest building with One World Trade Center. But when I asked "what's the second tallest?" the old Siri gave me the K2 mountain instead, which breaks the context I was going for.
The new Siri got the same first two answers right, and when I asked "what about in the U.S.?" I got the Willis Tower, so Siri thought I was still talking about the second tallest building in a new location. That's technically accurate but not what I was expecting.
Winner: Old Siri
New Siri vs Old Siri: Tell me about Paris
This round wasn't much of a contest. I started by saying "Tell me about Paris." and I got a very similar description from both assistants. And they both gave me the correct answer when I asked "how big is it?"
However, the old Siri got tripped up when I asked "How deep are the catacombs?" providing a more general answer. The new Siri gave me the depth of the Paris catacombs, and it even went on and gave me the hours.
New Siri vs Old Siri: Sports
I ask sports questions all the time to Siri so I thought this would be an easy one for both new and old. It was not.
I started by asking the Old Siri, "Who do the New York Jets play this weekend?" And I got a correct response. However, Siri got stumped when I asked "Who is favored to win?" I just got a bunch of Safari links to click off to.
The new Siri had its head in the game. I got the odds for the game as well as the over-under. But when I asked "Who is the quarterback?" Siri didn't supply an answer.
Winner: New Siri
New Siri vs Old Siri: Driving to the moon
For the next round, I wanted to see how well the new Siri could hold context when discussing the moon. I started asking "How far away is the moon?" followed by "How long would it take to drive there?"
The old Siri fell flat on the second question, saying it historically takes three days to reach the moon. Not while you're driving!
The new Siri got my gist, telling me it would take six months to drive there by car at 60 mph. It pulled the answer from Howstuffworks.com.
Winner: New Siri
New Siri vs Old Siri: How do I record my screen?
One of the benefits of the new Siri is that it can draw on large language models that have been trained on Apple support documentation to answer thousands of questions on Apple products.
So I put this capability to the test by asking "How do I record my screen?" The old Siri just gave me a bunch of web links, while the new Siri gave me step-by-step instructions, which is super helpful. Anyone who doubles as the IT person for this family will really appreciate sharing this feature with everyone — so they get fewer calls themselves.
Winner: New Siri
New Siri vs Old Siri: What's the weather?
You can use the old and new Siri to ask about the weather and get similar responses — even if you switch cities. But the new Siri is still a bit better.
I asked "What's the weather this weekend?" followed by "When's the next chance of rain?" and the new and old Siri both gave helpful responses. Both Siris also had no problem giving the 10-day forecast when I switched to Chicago.
But I like the new Siri a bit more in this round. First, you get more displayed info like the icon for the amount of sun or clouds to expect, and you'll see the wind direction and speed, too.
Winner: New Siri
Bottom line
The new Siri is simply better at maintaining context and providing more accurate info when you have follow-up questions. I also appreciate that it can continue listening without you have to press a button on your iPhone. Siri certainly has a long way to go to fulfill all of its Apple Intelligence promise, but it's off to a good start.
More from Tom's Guide
Sign up to get the BEST of Tom's Guide direct to your inbox.
Here at Tom’s Guide our expert editors are committed to bringing you the best news, reviews and guides to help you stay informed and ahead of the curve!
Mark Spoonauer is the global editor in chief of Tom's Guide and has covered technology for over 20 years. In addition to overseeing the direction of Tom's Guide, Mark specializes in covering all things mobile, having reviewed dozens of smartphones and other gadgets. He has spoken at key industry events and appears regularly on TV to discuss the latest trends, including Cheddar, Fox Business and other outlets. Mark was previously editor in chief of Laptop Mag, and his work has appeared in Wired, Popular Science and Inc. Follow him on Twitter at @mspoonauer.
-
Matt2h It's kind of a nonsensical question to ask, how long would it take to drive to the moon? I guess you could say suppose you were traveling about as fast as a car on the highway goes but obviously it's not even possible.Reply