This might sound underwhelming after generating songs and video clips from text prompts. But your smartphone almost certainly has more AI built into it than you realise, and most of it is designed to be so seamless that you never notice it's there. That's a completely different approach to AI than anything you've explored so far in this programme, and it's worth paying attention to, because it's how most people will actually encounter AI in their daily lives.
The chatbots and creative tools you've been using are things you go to deliberately. You open a browser, navigate to a website, type a prompt, and wait for a result. On-device AI is the opposite. It's woven into the operating system itself, quietly working in the background or appearing as an option at the exact moment you might need it. When you select text in a message and the phone offers to rewrite it for you, that's AI. When you point your camera at a restaurant menu in another language and see the translation overlaid in real time, that's AI. When you circle something on your screen and your phone instantly searches for it, that's AI too.
Apple, Google, and Samsung have all invested heavily in building AI features directly into their devices, and the pace of new additions is picking up. Many of these features run on-device, meaning your data stays on your phone rather than being sent to a server. That has real implications for privacy that we'll explore further in Thing 17.
This Thing is about finding the AI that's already in your pocket and working out which features are actually useful versus which are clever but forgettable.
What's built into your phone
The specific features available to you depend on your phone, its age, and its operating system. Rather than trying to cover everything (the feature lists change with every software update), here's a guided tour of the main AI capabilities on each platform as of early 2026.
The bigger picture: AI as infrastructure
There's something worth pausing on about on-device AI.
Every other tool you've used in this programme required you to make a conscious decision to engage with AI. You chose to open ChatGPT, to navigate to Suno, to sign up for Pika. On-device AI is different. It's there whether you choose it or not, built into the software you already use. You might have been using AI features for months without realising it.
This raises a question that doesn't have a single right answer: is invisible AI a good thing? If your phone can quietly remove background noise from a call, translate a sign while you're travelling, or clean up a photo without you needing to learn a new tool, that's a genuine improvement to everyday life. But there's also a case that people should know when AI is involved in what they're seeing or doing, so they can make informed decisions about trusting the output.
Both perspectives are reasonable, and we'll return to questions about transparency and trust later in the programme. For now, the important thing is simply to be aware of what's happening on your device, so you can make conscious choices about which features you use and which you don't.
Resources to explore
Apple's guide to what's available across iPhone, iPad, and Mac, with details on compatible devices and privacy architecture.
Step-by-step guide to using the system-wide writing features on iPhone. Apple Support.
Overview of AI capabilities built into Pixel phones, including Magic Eraser, Circle to Search, and Gemini.
Samsung's hub for Galaxy AI features including Live Translate, Note Assist, and generative photo editing.
Google's latest update on Circle to Search capabilities, including multi-object recognition.
A broader look at where on-device AI is heading and why it matters. Perficient.
Activity: the AI already in your pocket
You're going to spend half an hour exploring AI features that are already built into your phone, document what you find, and create a short visual guide that someone else could follow. This is one of the few activities in the programme that requires no new accounts, no free tiers to navigate, and no web tools at all.
- Go on an AI safari. Work through the device-specific instructions below that match your phone. Try each feature with something personal and take a screenshot of each one in action.
- Pick your top three. From everything you've tried, choose the three features that impressed you most, surprised you most, or that you could see yourself using regularly.
- Create your guide. Write a short guide called "Three AI features already on my phone," as if explaining them to a colleague who hasn't discovered them yet.
- Reflect. Add a short reflection on what surprised you, how on-device AI feels different from the web-based tools you've been using, and any concerns about AI being built into your phone at the OS level.
Before you start
Make sure your phone's operating system is up to date. Many AI features arrive through software updates, so if you've been putting off that "Update Available" notification, now's the time. On iPhone, go to Settings > General > Software Update. On Android, go to Settings > System > Software Update (the exact path varies by manufacturer).
If you have an iPhone 15 Pro or later, check that Apple Intelligence is enabled: Settings > Apple Intelligence & Siri. The on-device AI models may need to download the first time you enable the feature, which takes a few minutes on Wi-Fi.
Go on an AI safari
Work through the list below that matches your phone. For each feature, try it out with something personal and take a screenshot of each one in action.
Create your guide
Write a short guide called "Three AI features already on my phone." Imagine you're explaining these features to a colleague who hasn't discovered them yet. For each of the three features, include:
- What it's called and where to find it
- A screenshot showing it in action
- A plain-language explanation of what it does (assume your colleague hasn't used any AI features on their phone)
- Your honest assessment: is this actually useful in everyday life, or is it more of a novelty?
Keep the tone practical and conversational. The goal is something your colleague could read in five minutes and immediately go and try themselves.
Your output
A screenshot-illustrated guide containing:
- Your top three AI features, each with a screenshot, explanation, and honest assessment
- A short reflection (a couple of paragraphs) covering: how many features did you already know about? How does on-device AI feel different from the web-based tools you've been using? Do you have any concerns about AI being built into your phone at the operating system level?
Things to notice
On-device AI features tend to respond almost instantly, because they're not sending your data to a server and waiting for a response. When you use Writing Tools on an iPhone or Circle to Search on a Pixel, the result appears in a fraction of a second. Compare this to the seconds (or minutes) you waited for AI video generation or deep research. That speed comes from running smaller, more specialised AI models directly on the phone's processor rather than the large general-purpose models that power chatbots.
The results are also more constrained. On-device AI doesn't try to be a general-purpose intelligence. It does a small number of things, each one well-defined: rewrite this text, identify this object, remove this element from a photo. You won't get the open-ended conversational capability of ChatGPT or Claude. The trade-off is reliability; these features tend to be more consistent than general-purpose AI, because they're optimised for specific tasks.
Most of what happens stays on your phone, too. Apple is particularly explicit about this; their privacy model is a central part of Apple Intelligence. Google processes more in the cloud but has been moving toward on-device processing for features like Circle to Search and call screening. Samsung's Knox security framework provides additional data isolation. This is a meaningfully different privacy model from typing your thoughts into a web-based chatbot.
If you're doing this activity alongside someone with a different phone, you may notice real differences in what's available. This is partly about hardware generations and partly about the different bets each company is making. Apple prioritises privacy and system-level integration. Google prioritises search and cross-app intelligence. Samsung combines both approaches with its own additions. There's no single "best"; it depends what you value.
Why this matters
You've spent thirteen Things going out to find AI: signing up for services, typing prompts into web tools. Here, the AI came to you. It was already on your phone, waiting to be noticed.
This is increasingly how most people will interact with AI. Your phone's photo editor suggests removing a distracting object. Your email app offers to summarise a long thread. Your keyboard offers to rewrite your message in a more professional tone. These are small, practical interventions that accumulate over time.
It changes the question from "should I use AI?" to "which AI features am I already using, and am I making deliberate choices about them?" AI is moving from being a separate category of tool to being a layer that runs underneath everything. The more aware you are of that layer, the better equipped you are to use it thoughtfully.
Claim your Open Badge
Submit your screenshot-illustrated guide covering three AI features, your written explanations and assessments of each, and your reflection on the experience as evidence for your Thing 14 badge via cred.scot.
Submit your screenshot-illustrated guide and reflection as evidence to claim this badge via cred.scot.
Claim now