Artificial intelligence: these are the smartest smartphones

At trade fairs like the Mobile World Congress and IFA, Huawei, Samsung and Qualcomm are the main companies that focus on AI. There's the new Kirin 980 with not one, but two NPUs from Huawei's chip subsidiary Hisilicon, Bixby 2.0 from Samsung and Qualcomm's AI Engine. You can also throw in Apple's new bionic processor and Google's Pixel Visual Core into that mix. These are currently the three important areas of application for AI on smartphones:

• Photo, video and image recognition: detecting objects in front of the camera, adjusting settings for photos and videos, and cataloging images based on their content. These are things that can easily be optimized with AI tools, such as machine learning chips or algorithms. That's precisely what Google, Huawei and Apple do with their AI solutions. Qualcomm's AI Engine is also optimized for this kind of processing.
• Language and translation: machine learning is an excellent way to recognize patterns in spoken language. It's used to recognize a speakers' commands, even when at a distance or with the presence of background noise. It helps Alexa, Siri, Cortana and co. communicate better. Translation services are also important in the field of communication. Microsoft's translator and DeepL rely heavily on AI to yield more natural translation results.
• Performance and battery life: making tasks that utilize AI features less battery-draining and more efficient overall. The more AI tools are used sensibly, the more potential there is for improving battery life.

Of course, this is only the current state of AI functions. AI is growing rapidly in the world of smartphones, and in the future we'll see many more applications. Development is progressing and we're still just at the beginning. Yet, even today there are exciting AI functions you can enjoy, if you have the right smartphone.

The most 'intelligent' smartphones at a glance

Huawei Mate 20 (Pro)

It's the Mate 20 Pro and it's smart alright. Huawei's new flagship smartphone is the first to feature the Kirin 980, as well as pack two NPUs. The device benefits immensely from this. The features alone are very useful: scene recognition on the camera, facial recognition, photo sorting and direct translation on the device without the need for an online connection. Above all, it's the performance improvement compared to the Kirin 970 that really distinguishes the new chip. Oh and by the way, the Kirin 980 is also available on the standard version of the Mate 20.

• Huawei Mate 20 Pro: neck and neck with Apple and Samsung
• Mate 20 hands-on: Huawei's attempt to strike a happy medium

Google Pixel 3 (XL)

Google recently released its Pixel 3 along with the larger Pixel 3 XL. Artificial intelligence plays an important role for both devices in two areas: Google's language assistant is always in the process of learning more, and the camera, which only has one lens, is extremely smart. The elaborate software provides great results with Google relying on the latest Snapdragon processor.

• Pixel 3 XL hands-on: a few tweaks and a steep asking price

Huawei P20 (Pro)

Aside from the brand's new Mate line, the Huawei P20 and P20 Pro are popular and up-to-date smartphones. The Kirin 970 was the first chip with its own NPU in a smartphone, and its functions don't differ much from its successor, the Kirin 980. Although the newer chipset is much faster, it doesn't differ that much from its "old" processor. In concrete terms, this means that the AI helps create great photos, while the NPU offers many advantages in terms of image search and power consumption.

• Huawei P20: New look and AI for the P series
• Huawei P20 Pro: the three-eyed monster

In n In the Huawei P20 Pro, the AI is mainly used for the camera. / © NextPit

Honor 10

The Huawei subsidiary brand Honor presented this price-performance darling this spring. The Honor 10 is basically a P20 Pro with just two cameras - it has the same processor and similar features. Honor is even smarter when it comes to photos, since you can switch off the AI effects afterwards, if they don't fit the picture. With over 500 scenarios and 22 categories, the camera recognizes virtually anything in front of its lens. The Honor 10 even goes as far as splitting the picture into several areas and then edits each section individually via the software. It's so fast that you barely notice it.

• Honor 10 review: OnePlus has something to worry about

Apple iPhone XS

With the A12 Bionic in the iPhone XS, Apple prides itself on having the fastest AI processor in any smartphone. The octacore chipset computes five trillion arithmetic operations per second in the NPU alone, which is an impressive figure. Apple specializes in AR apps, where it's important to recognize the environment as quickly and accurately as possible - no device does this as well as the iPhone XS. Apple has also made machine learning technology available to external developers through the Core ML platform.

• 48 hours with iPhone XS: back to square one

Samsung Galaxy S9 / S9+

Samsung doesn't have an AI unit in its processor, but still relies on artificial intelligence through its language assistant Bixby. Bixby has its own button on the Galaxy S9 and with Bixby 2.0 there are more and more features and possibilities. You can even integrate the assistant into a smart home network with kitchen appliances and an alarm system.

• Samsung Galaxy S9 review: love and hate

Samsung Galaxy Note9

The Samsung Galaxy Note 9 is pretty much a Galaxy S9 on steroids: it's bigger, more powerful and comes with a unique stylus. But both smartphones use the Exynos 9810, which has to get by without its own AI unit and the Bixby button. Even though Samsung's assistant is developing comparatively slowly, the market leader will continue to rely on it in the future because Bixby will be integrated with other networked devices, such as the Family Hub super refrigerator.

• Galaxy Note9: the smartphone of your dreams for a steep price

What do you think of artificial intelligence? Where do you think it's headed? Let us know in the comments below.