Are smartphones becoming smarter with on-device AI? | Explained

The smartphone industry is shifting from cloud-based AI to on-device processing, utilizing dedicated Neural Processing Units (NPUs) to run machine learning models locally. This transition enables faster response times, improved privacy, and offline functionality for generative AI features.
Why it matters
This technological shift fundamentally changes how users interact with mobile devices, prioritizing local data security and performance over constant server connectivity.
The transition of Artificial Intelligence (AI) from the cloud to the local device , often called on-device AI, is arguably the most significant shift in the smartphone industry since the inception of the app ecosystem. For over a decade, consumer AI has been synonymous with the cloud: when you asked a question to an assistant, your data travelled across the globe to a massive data centre, was processed, and then sent back to your device. Now, users are witnessing a seismic shift. Intelligence is moving from remote servers to the silicon residing in their handheld devices.
On-device AI refers to the processing of machine learning algorithms and generative AI models directly on the smartphone’s local hardware. Instead of relying on a constant connection to a remote server, the device performs the computation using dedicated silicon, primarily Neural Processing Units (NPUs).
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in