Mobile is the personal computing platform now. Once the cool new technology on the block, mobile has now become the ’new normal’.

And it’s here to stay.

10 humans standing as an illustration. Nine are blue and holding a smartphone, one is yellow and isn't holding anything. | Halcyon Mobile

By 2025, almost 9 in 10 people will have a mobile device in their pocket in North America and Europe.

10 humans standing as an illustration. Seven are blue and holding a smartphone, Three are yellow and aren't holding anything. | Halcyon Mobile

And the rest of the world is catching up. By 2025, about 7 in 10 people will have a mobile device worldwide1.

And soon enough, everyone will have an internet connection through a mobile device. This has mostly exploded thanks to smartphones.

By 2025, around 8 people in 10 will have a mobile internet connection2.
Wow! Remember when all we had was dial-up?

Mobile internet penetration Europe

Mobile internet penetration North America

Mobile internet penetration worldwide

What are people doing with their smartphones? All sorts of things3.

Abstract illustrated humans around an oversized smartphone they seem to work on. One has his back towards us, tapping on the screen. The woman is sitting on top of the smartphone. Beside her is a speech bubble, where she says: "That's... a lot." | Halcyon Mobile

And people would rather stay on mobile than switch back to a laptop or desktop computer. In fact, they spend more time on mobile than on any other device4.

Daily hours spent with Digital Media per Adult User (USA)

Mobile

Desktop / Laptop

Other connected devices

In fact, most people can’t wait to try out new mobile products.

Engagement patterns are strongly skewed towards early adoption5.
Smartphone user segmentation by engagement pattern (% of total smartphone users)

engagement patterns of mobile phones based on geographic areas. America, Europe and Middle East and Northern Africa are skewed towards the right, with more power users than average. Areas in development or underdeveloped mainly use their mobile phones for calls and messaging. | Halcyon Mobile

Afficionados use their phones for everything. They show high engagement across all areas.

Pragmatists use their smartphones for most things, but are still experimenting with lifestyle, digital commerce and financial services use cases.

Networkers use their phones to communicate, socialize, browse the internet and for apps, and sometimes for entertainment.

Talkers use their mobile phones almost exclusively to make a voice call or send an SMS.

And it’s not just about our lives and businesses going mobile6. How mobile shapes our day-to-day is getting transformed by new technologies.

We think the next chapter in mobile is AI.

And it’s not just us being excited about AI’s potential in tech.

The number of AI startups and VC investments made in AI have exploded in the last 5 years8.

+113%

Increase of active AI startups in the US from January 2015 to January 2018

+350%

Increase of venture funding for AI startups in the US from January 2015 to January 2017

In the last years, the main applications of deep learning have been defined7.

Deep learning is transforming how software is working, by changing the focus from code to data. Deep learning algorithms learn from examples, not from code. It has applications in:

Eye illustration | Halcyon Mobile

Computer vision

Speech bubble illustration | Halcyon Mobile

Natural language processing

Patterned pixels in a circle | Halcyon Mobile

Pattern recognition applied to sensory data

Wait! What’s deep learning again?

Machine learning refers to the ability of a machine to learn using large data sets instead of hard-coded rules. Deep learning is a machine learning technique, loosely modelled on our brain structure. The algorithm performs mathematical computations on input data, through several layers of “neurons” (where each neuron makes a specific calculation), until they return the desired outputs7.

Illustration of half an eye and half a camera sensor, standing in for a smart camera | Halcyon Mobile

The camera becomes a smart eye, not only capable of seeing in different lighting settings, but also recognizing what it sees (people, writing, objects)

An illustrated smartwatch with half a heart coming out beside it. | Halcyon Mobile

Paired with a SMART WEARABLE, you can get a new level of body awareness.

Abstract illustration of a woman sitting in front of a human-sized smartphone tapping on the screen. | Halcyon Mobile

YOUR SMARTPHONE will move beyond sensors and get artificial “senses”, capable of reaching human performance in some areas and even going beyond human performance in others.

Illustration of half a microphone and an ear, standing in for a smart microphone | Halcyon Mobile

The microphone becomes a smart ear, able to recognize what it hears (your voice, what you say, surrounding sounds).

Illustration abstracting the gyroscope visual, as stand in for smart sensor data reading | Halcyon Mobile

The acceloremeter and giroscope give you a smart sixth sense of motion.

Abstract illustration of a woman sitting in front of a human-sized smartphone tapping on the screen. | Halcyon Mobile

YOUR SMARTPHONE will move beyond sensors and get artificial “senses”, capable of reaching human performance in some areas and even going beyond human performance in others.

Illustration of half an eye and half a camera sensor, standing in for a smart camera | Halcyon Mobile

The camera becomes a smart eye, not only capable of seeing in different lighting settings, but also recognizing what it sees (people, writing, objects)

Illustration of half a microphone and an ear, standing in for a smart microphone | Halcyon Mobile

The microphone becomes a smart ear, able to recognize what it hears (your voice, what you say, surrounding sounds).

Illustration abstracting the gyroscope visual, as stand in for smart sensor data reading | Halcyon Mobile

The acceloremeter and giroscope give you a smart sixth sense of motion.

An illustrated smartwatch with half a heart coming out beside it. | Halcyon Mobile

Paired with a SMART WEARABLE, you can get a new level of body awareness.

There are approx. 2.71 billion smartphones in use worldwide in 20199, waiting to be upgraded with smart software to become capable of using deep learning for specific tasks. As deep learning gets deployed at scale, it will have a massive impact, creating a new generation of applications.

Here’s just a selection of the smart apps you may use or will end up using in the following years.

Smart apps based on computer vision

An alarm app so you can use a spare camera as a security system10

A camera update that can help you take photos in the dark11

An app that can tell you what flower you’re looking at12

An app that can recognize plant diseases13

An app that can recognize skin cancer14

An app that can count your exercise reps and guide you to keep the right form15

An app that can recognize cells or chromosomes when connected to a microscope16

An app that can count cars, people or animals passing by a certain area17

An app that can detect when someone has fallen and notify you18

An app that can sort cucumbers19

An app that lets you control your phone or mobility aids by using just your eyes20,21

An app that can help you draw22

An app that can live translate sign language23

An app that can guide you while assembling your furniture24

Smart apps based on NLP

An app that can recognize what you say, reliably, in any language25

An app that can live translate from one language to another26

An app that can detect your emotions from the tone of your voice27

An app that can detect your emotions from your words28

An app that can summarize any article29

An app that can help you write better30

An app that recognizes the music you’re playing and acts as an accompanist34

Smart apps based on pattern recognition in movement data sequences

An app that can recognize movement activities31

An app that can recognize sleep patterns32

An app that can recognize you based on your typing patterns33

An app that can help diagnose neurological diseases from body movement patterns35

An app that learns how your dog usually acts and can tell you when it’s sick based on its behavior36

Smart apps based on wearable data collection

A wearable device that can help you manage hand tremors37

An app that can help you recognize patterns of heart disease38

An app that can help you recognize patterns of stress39

An app that can help you detect in realtime one’s fertility window40

And many more apps we haven’t thought of yet41.

[1] GSMA – The Mobile Economy Global Report 2018

[2] International Telecommunications Union – Global Information and Communications Technology developments, 2001-2018

[3] Most popular Apple App Store categories in September 2018, by share of available apps

[4] Kleiner Perkins 2018 Internet Trends

[5] World Economic Forum – Digital Transformation Initiative, In collaboration with Accenture – Unlocking $100 Trillion for Business and Society from Digital Transformation, May 2018

[6] Timeline of artificial intelligence

[7] Want to know how Deep Learning works? Here’s a quick guide for everyone.

[8] Artificial Intelligence index, 2018 Annual Report

[9] Number of smartphone users worldwide

[10] Salient Eye

[11] Google Night Sight

[12] Picture This

[13] Pestoz

[14] Skin Vision

[15] Mira Rehab

[16] Image Scan

[17] Tutorial: Making road traffic counting app based on computer vision and OpenCV

[18] Fall Safety App

[19] How a Japanese cucumber farmer is using deep learning and TensorFlow

[20] EVA Facial Mouse

[21] EyeTech Digital Systems

[22] $1.1 billion Airtable is turning admins into app developers

[23] Augmented Reality App Can Translate Sign Language Into Spoken English, and Vice Versa

[24] This conceptual AR app helps you build Ikea furniture

[25] Sure Pet Care – Animo

[26] Google Assistant will soon detect what language you’re speaking in

[27] Google Translate adds real-time translations for 13 new languages

[28] Emotion AI overview

[29] Natural language

[30] Text summarization with TensorFlow

[31] Grammarly

[32] How to track exercises in mobile recognition app. Our Case-Study

[33] Sleep Score

[34] Typing DNA

[35] Metronaut

[36] AI can spot signs of Alzheimer’s before your family does

[37] Project Emma

[38] 4 Ways In Which AI Is Revolutionizing Cardiac Care

[39] This Phone App Knows If You’re Depressed

[40] Hello Clue

Deep learning in mobile has an endless range of applicabilities. It will be really interesting to watch how it all unfolds and we’re excited to be a part of it.

Illustration of an excited male waving his hands around with a speech bubble that says: "What an amazing time to be alive!" | Halcyon Mobile

Join us in discovering the AI-powered mobile world!

Want to be up to date with AI applications in products? Sign up for our newsletter below.