Augmented reality (AR) apps can enhance your experience of the real world by overlaying digital information, graphics, or sounds on your device's camera view. But how can you design AR apps that protect privacy and avoid intrusive content? Here are some tips to consider when creating AR apps that respect users' rights and preferences.
Before you start developing your AR app, you need to have a clear idea of what problem you are solving, who your target audience is, and what features you need to provide. This will help you avoid adding unnecessary or unwanted content that could compromise privacy or annoy users. For example, if your app is meant to help users find nearby restaurants, you don't need to access their contacts, messages, or photos. You should also limit the scope of your app to the relevant context and location, and avoid overlaying content that could interfere with users' safety, navigation, or attention.
This is the simple AR case where a camera is required and data about a person's environment and movements and gait is required to log into an account or offer some other service. Gait is your unique shape and movements that are nowadays used as authenticators for various purposes. This data can also be used for creating ML models that can recognize your movements to interact with a virtual reality. It can be stored on centralized servers and if there was a breach (internal or external, see Snowden) and authoritarian groups got a hold of such personal data, they could use city cameras and other devices to track and identify people based on their gait.
One of the most important aspects of designing AR apps that protect privacy is to request permission and consent from users before accessing or using their data, camera, microphone, or location. You should explain why you need these permissions, how you will use them, and how users can revoke them if they change their mind. You should also respect users' choices and preferences, and provide them with options to customize their experience, such as adjusting the level of detail, brightness, or volume of the AR content. You should also avoid collecting or storing any personal or sensitive data that is not necessary for your app's functionality.
Consent is becoming less and less of an issue as GDPR starts enforcing very real standards and retroactively punishing big corporations.
Another way to design AR apps that protect privacy is to use encryption and authentication to secure the data transmission and storage between your app and your server. Encryption means that the data is converted into a code that can only be read by authorized parties, while authentication means that the parties have to prove their identity before accessing the data. This will help prevent hackers, third parties, or malicious actors from intercepting, modifying, or stealing your users' data. You should also use secure protocols, such as HTTPS, SSL, or TLS, to ensure that your app's communication is encrypted and authenticated.
Simply encrypting and decrypting gait data is not sufficient. We must use some form of Federated Learning or some local pre-processing that would allow gaits to be shared in a totally or statistically private way, while allowing the features and services to continue operating as before.
Finally, you should test and update your app regularly to ensure that it works properly, that it does not contain any bugs or vulnerabilities, and that it complies with the latest standards and regulations. You should also monitor your app's performance, feedback, and usage to identify any issues or improvements that you need to make. You should also inform your users about any changes or updates that you make to your app, especially if they affect their privacy or consent. By testing and updating your app regularly, you can ensure that it is secure, reliable, and user-friendly.