We use Google Analytics to measure site usage. This requires a cookie. Support our analysis by clicking Accept.

TL;DR;

Dr. Iain Keaney talks about contact tracing apps and privacy rights. And how decentralisation could be the answer.
Guest contributor on our #AIFightsBack webinar series was Dr. Iain Keaney, Data Scientist and Founder of Skellig.ai. Iain took on the world of contact tracing apps and discussed how both governments and companies are fighting the pandemic while mitigating the risk to personal privacy. He pointed to the adoption levels of different countries with China having the largest-scale adoption of contact tracing apps in the world and Iceland next at 40%.

Guest contributor on our #AIFightsBack webinar series was Dr. Iain Keaney, Data Scientist and Founder of Skellig.ai. Iain took on the world of contact tracing apps and discussed how both governments and companies are fighting the pandemic while mitigating the risk to personal privacy. He pointed to the adoption levels of different countries with China having the largest-scale adoption of contact tracing apps in the world and Iceland next at 40%.

Norway in contrast has recently suspended its virus-tracing app due to privacy concerns and low adoption. Iain also noted that figures relating to adoption and efficacy vary greatly with a minimum adoption of 20% reported to be required to have any kind of impact on the pandemic.

So where does this leave users and their general mistrust in the technology and public safety?

The answer: Decentralised contact tracing apps

According to Iain, decentralised contact tracing apps are the answer to reaping the benefits of the technology to fight the pandemic and to protecting our right to privacy. There is no mass data collection or location tracking and this is the main principle behind the Google/Apple API collaboration. This approach is based on the exchange of randomised key codes from the users mobile phone and if COVID 19 symptoms occur, the user notifies the app and any matches will be alerted.

Video

Decentralised Machine learning

This is all based on the concept of Federated Learning (collaborative learning), which is a decentralised machine learning (ML) technique that gives us a much greater degree of data privacy. It also produces a greater degree of personalised AI, a good example of this is the predictive text on our mobile phones. Essentially, AI models learn from the user and customise their experience. If you combine Personalised AI with Global AI the entire systems learns and improves as one unit. Adding encryption and anonymising data before it leave the users mobile device prevents any re-identification and protects privacy. This is the basic idea anyway.

In terms, of putting trust back into contract tracing apps, decentralised ML is the way forward and why big companies like Google and Apple have invested heavily it but there is a long way to go.

For more information or to get in contact visit www.skellig.ai

#AIFightsBack