After a long period of minor advances in innovation, smartphones deserve close attention again.
Unless you’ve been willing to put up with the relatively poor battery life of foldables, smartphones have become boring slabs of glass. From one Apple iPhone or Samsung Galaxy S series device to the next, there have been few meaningful changes in at least half a decade.
Manufacturers such as Apple, Samsung and Huawei found that the only way they could make their products stand out was to design unique camera bumps — in Apple’s case, a distinctive triangle of lenses; in Samsung’s, a vertical lens arrangement. It’s basically the only way to see at a glance what phone someone is using.
So, what is changing to make these devices turn heads again?
For one thing, folding phones such as Samsung’s Galaxy Z series have reached a level of maturity where they’re arguably ready for broader consumer adoption (bar their stubbornly high prices).
Samsung’s new Galaxy Z Fold7 device — the seventh generation in the series — is a big step forward for the form factor. Not only has the hinge design been largely perfected, but advances in chip and battery technologies have allowed the company to slim the device down to such an extent that when it’s folded, it’s the same thickness as its flagship Galaxy S25 series that uses the traditional “candy bar” (nonfolding) form factor.
Big changes are coming soon to the iPhone too. In a few weeks, Apple will take the wraps off a new iPhone category: alongside the expected 17, 17 Pro and 17 Pro Max, the iPhone 17 Air (the name is not official yet) will make its debut. It will be significantly thinner than iPhone users have grown accustomed to, but will come with trade-offs. (Apple, by the way, is also believed to be working on a folding phone, with a rumoured launch date of 2026.)
The impending launch of the Air comes months after Samsung unveiled an ultra-thin phone of its own, the Galaxy S25 Edge. But thin devices are more difficult to keep cool inside, leading to thermal throttling, and their batteries are smaller because there’s less space, meaning users struggle to get through a full day without an energy top-up.
Apple and Qualcomm (a Samsung chip supplier), however, are betting that advances in semiconductor manufacturing — the companies are gearing up to move to an advanced, 2-nanometre node next year — will deliver devices that use considerably less power while offering the same (or better) performance. It’s Intel co-founder Gordon Moore’s famous law in action.
This year’s first iPhone Air will use a 3nm node (as will the other iPhone 17 models, for the third year in a row). But Apple is expected to move to a 2nm node with partner Taiwan Semiconductor Manufacturing Co in the iPhone 18 lineup next year, which should help the second-generation Air’s battery life considerably.
Hardware is, however, only one reason smartphones are becoming exciting again. The second is AI.
Smartphones are already powerful enough to run some of the simpler language models — and will only get better at this every year
Companies such as OpenAI (which recently hired iPhone designer Jony Ive) and Meta Platforms are putting considerable effort into building AI-powered devices that could supplant the smartphone. Despite this, phones will remain the main way people interact with large language models, at least until the end of this decade.
But as demand grows from consumers for greater privacy (and less mining of their personal data for targeted advertising), on-device AI processing will become increasingly important. Today, only ultra-modern phones and PCs have the processing grunt to do that. Smartwatches and smart glasses won’t have the same power, meaning they will have to be paired with a smartphone, or the processing will have to take place in the cloud for more advanced queries, with the privacy concerns that brings.
AI is ushering in an exciting new era of computing — one whose impact will be at least as profound (for jobs, productivity and economic growth) as the internet has been. The impact on the media industry alone will be as significant as the move from print to online.
But it will take years of advances in silicon, in line with Moore’s law, before lightweight wearables are powerful enough to run high-quality language models on-device that can interact with you (and your environment) without first uploading everything into the cloud for processing.
Smartphones, on the other hand, are already powerful enough to run some of the simpler language models — and will only get better at this every year. Ergo, they have real staying power, even if Ive builds something incredible in his new lab at OpenAI.
McLeod is editor of TechCentral
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.