Apple and Google release marks 'watershed moment' for contact-tracing apps

Apple and Google have released a software tool that will make it possible for nations to release coronavirus contact-tracing apps that adopt the firms' privacy-centric model.

It offers developers access to added Bluetooth functionality to solve a problem existing apps have of iPhones sometimes failing to detect each other.

Android and iOS device owners will have to carry out system upgrades.

But some countries - including the UK - are pursuing a different approach.

"The release of these APIs [application programming interfaces] along with the operating-system updates will be a watershed moment for the development and adoption of proximity-tracing apps," said Marcel Salathé, an epidemiologist at the Swiss research institute EPFL.

He added that apps that adopted the protocol should be able to be made "interoperable" - meaning that citizens can continue to be contact-traced as they cross from one region and/or country to another. That could potentially help reduce travel restrictions imposed because of the virus - at least for those using the apps involved.

Apple and Google said public health agencies from 22 countries had already asked to test the system.

The app was not "a silver bullet" - but "user adoption is key to success and we believe that these strong privacy protections are also the best way to encourage use".

On-device matching
Contact-tracing apps are designed to automatically log when two people come into proximity to each other for a significant amount of time.

If one is later diagnosed with the coronavirus, the other can be given an alert, which might suggest they self-isolate and/or request a medical test of their own.

But the authorities believe adoption has been hampered by two factors:

concerns the technology poses a privacy risk
restrictions Apple places on third-party apps' use of Bluetooth in the background
In theory, the new system should address both these issues.

Its "decentralised" approach locates contact-matching on devices themselves rather than a centrally controlled computer server.

And this aims to cut the risk of either hackers or the authorities using the database of who met whom and for how long for other purposes.

But the UK's NHS and its counterparts in France, Norway and India say the centralised approach gives them greater insight, making it easier to tweak the risk model that decides who receives which type of alert.