New Era of AI Smartphones: Embedded AI on the Move

After years of flagging sales, mobile phone makers are betting big on artificial intelligence. A raft of recent announcements suggests embedded AI is set to dramatically change how we use our handsets, interact with apps, and even drive our cars.

With capabilities like conversational commerce, advanced text generation, and in-image search rapidly taking shape, the age of the simple ‘smart’ phone may be coming to an end.

What does it mean for end users? Can AI phones really live up to the hype?

A flurry of embedded AI announcements from Samsung and others suggests 2024 will be the year of the AI smartphone.

In this article, Techopedia looks at the latest embedded AI trends and talks to the experts about what’s coming down the pipe.

Key Takeaways

  • The AI onslaught has upended software, search and SaaS. Now it’s coming for mobile.
  • Handset manufacturers hope embedded AI can revive interest in the category and shorten upgrade cycles.
  • Some of the features currently on offer have a gimmicky feel, but analysts believe a new era in handheld computing could be on the horizon.
  • Addressing concerns about privacy and security is something everyone in the smartphone AI ecosystem will need to consider.

Why AI Smartphones Are Breaking the News

AI dominated the agenda at February’s Mobile World Congress (MWC) event in Barcelona, where big-name tech brands like Samsung, Qualcomm and Google promised a new era of AI in the palm of your hand.

Advertisements

Samsung led the pack, launching its Galaxy S24 line of AI-enhanced phones in January.

Xiaomi and Honor showcased the generative AI capabilities of their latest handsets, while Qualcomm and Mediatek both launched chips that support on-device AI.

While many upscale smartphones have had AI features for years, most of the intelligence happened in the cloud.

The new generation of AI smartphones comes with hardware, storage and OS capabilities that allow the computational heavy lifting to happen on the phone itself, making the experience faster, more responsive, and even more personalized.

The Biggest Embedded AI Announcements So Far in 2024

Samsung AI Smartphone S24

Samsung’s S24 series launch gave the Korean tech giant claim to the title of ‘first AI smartphone’ back in January.

Using Google’s Gemini AI and Imagen 2-based text-to-image tool, the three S24 series phones boast an on-device LLM that’s integrated with Android 14 OS.

AI capabilities have been added to Samsung-native apps, which include enhanced photo-editing capabilities through Imagen 2.

Xiaomi 14

Unveiled last October, the now-launched Xiaomi 14 received a lot of attention at MWC alongside its upscale sister phone, the Xiaomi 14 Ultra.

Both offer on-device generative AI through Quallcom’s Snapdragon 8 Gen 3 platform — the first phones to do so.

Highlight features include the ability to extend an image outside its original borders and let AI project what should fill the space.

You can also train it with images of your face to create ‘selfies’ that place you in novel or exotic settings.

Honor Magic 6

Also making its debut in Barcelona was the Honor Magic 6 Pro, which hands users an AI-driven navigation feature that allows the phone to be controlled with eye movements. A tracking feature follows where you’re looking and predicts where you’d like to go next.

The Chinese company even claims you can use the feature to control your car using eye movements alone.

Qualcomm

Running an AI model on any device requires a lot of processing power. Qualcomm debuted the first GenAI-ready mobile hardware late last year and launched new chips at MWC that offer built-in AI capabilities.

The Snapdragon X80 5G modem is designed for AI phones, tablets, PCs and other devices that run AI natively. While its dedicated AI processor promises improved data speeds, latency, and switching between wireless spectrums.

Google

Google arrived at MWC hoping to position Android as the go-to platform for mobile AI.

New capabilities for the operating system highlighted in Barcelona included a chat with Gemini, which enables users to access Gemini directly from inside the Google Messages app on any Android device.

Another feature coming soon will allow you to have long texts and noisy group chats summarized and read out while you’re driving.

AI Smartphones Momentum: Why Is This Happening Now?

According to data from the International Data Corporation (IDC) Worldwide Quarterly Mobile Phone Tracker, global smartphone sales declined 3.2% year over year to 1.17 billion units in 2023.

High-end phones reached a common level of quality and capability a few years back and differentiation is now difficult. Consumers often see the latest launches and go ‘meh.’ The hope is that embedded AI can kick start new innovations and give the sector a much-needed boost.

Is that realistic? Runar Bjørhovde, an analyst at Canalys, told Techopedia that the AI smartphone “super use case hasn’t arrived yet.”

He notes that smartphones have been running the equivalent of a small language model for years to support text auto-suggest and enhanced photo functions.

To move to the next level, he says, a smartphone needs to be fully “AI capable,” meaning it could host a full-fat LLM natively with no degradation of speed or performance — something the latest devices appear to make possible.

“When ChatGPT was launched, a lot of hardware manufacturers were caught completely off guard. Now they realize a paradigm shift is coming where AI could fuel new ways to monetize customers or help them differentiate.”

Canalys defines an AI smartphone as a handset capable of running a large language model (LLM) like ChatGPT natively, with hardware, storage, and operating system enhancements that enable the on-phone LLM to process a query quickly and return a rich result.

Even if current AI features — heavily weighted to image tweaking — have a ‘party trick’ quality, the research firm estimates that AI-capable smartphones will be 45% of the market by 2027.

AI-capable smartphones will be 45% of the market by 2027.

Anshu Goel, Lead Analyst, Digital Consumer at GSMA Intelligence, echoes that view.

He says that for smartphone users, “the value perception of advanced AI will likely rise as the tech slowly but surely permeates the entire gamut of mobile digital services.”

Goel adds that the GSMA’s own research shows almost half of users with plans to replace their handsets view AI capabilities in smartphones as “very or somewhat important.”

Security Issues: Embedded AI’s Biggest Worry

The arrival of AI on smartphones could be a gift to cybercriminals. In an interview with security consultant Frank Abagnale (memorably portrayed by Leonardo DiCaprio in Catch Me If You Can), Techopedia’s Linda Rosencrance captured his view that AI on smartphones could lead to more sophisticated social engineering scams.

Abagnale advised anyone to move away from password security and adopt an anonymized passkey to better protect their online activities. He said:

“For the past 20 years, I have sat on the Board of Advisors of a technology company. We developed the no-password technology known today as a passkey. The passkey has been adopted by the FIDO Alliance, Apple, Google, and Microsoft. Today, every device has passkey technology. This will eventually do away with usernames and passwords. This is long overdue.”

The Bottom Line

Will we look back and say 2024 was the year the smartphone died? The AI nirvana of anticipatory computing still seems a long way off. Many of the AI phone features being touted now have a gimmicky feel, and there’s a risk of AI fatigue if consumers come to believe that manufacturers are over-promising.

But there have also been intriguing innovations — Honor’s eye-tracking control interface, Google’s Circle to Search feature, and an app-free AI phone concept from Deutsche Telekom all point to bigger and better things. It’s just a question of when.

FAQs

What is AI in smartphones?

Which mobile phones have AI?

What is the new AI phone in 2024?

Do iPhones have AI in them?

Advertisements

Related Reading

Related Terms

Advertisements
Mark De Wolf
Tech Writer

Mark is a freelance tech journalist covering software, cybersecurity, and SaaS. His work has appeared in Dow Jones, The Telegraph, SC Magazine, Strategy, InfoWorld, Redshift, and The Startup. He graduated from the Ryerson University School of Journalism with honors where he studied under senior reporters from The New York Times, BBC, and Toronto Star, and paid his way through uni as a jobbing advertising copywriter. In addition, Mark has been an external communications advisor for tech startups and scale-ups, supporting them from launch to successful exit. Success stories include SignRequest (acquired by Box), Zeigo (acquired by Schneider Electric), Prevero (acquired…