Content area
Full Text
1. Introduction
In today’s age of technological revolution, it is extremely difficult for the law to anticipate potential data misuse and keep up with the societal impacts this causes. This is especially true for the exponential advancements in the scope and speed of artificial intelligence (AI).
One particular area of shortcoming that has been exposed by the rapid advancement of AI is the violation of privacy [1]. From retail loyalty cards, to workplace access cards, we are living in a world where modern surveillance has become common practice (Lyon, 2001). Furthermore, the data protection and privacy concerns for emerging technologies are growing every day, and this is further complicated by the vast amount of personal information available in the public domain coupled with the Internet of Things (IoT) and various social media collecting data about everything that consumers interact with.
Concerns around privacy of sensitive personal data are often not immediately obvious to consumers, as the understanding of the technology and data captured is not obvious. And in many cases, the law is not always able to protect the consumer as not all data is considered personally identifiable data. Although the issues around AI and the protection of personal data has gained more prevalence, there is no real consensus amongst governments, industry, advocacy groups and even academics as to how data privacy must be protected (Kuner et al., 2018).
The purpose of this paper is to provide a real world case study to illustrate the ethical considerations arising from the application of AI to everyday “anonymous” data, in this case, home energy data. We show how AI has the potential for energy companies to reliably identify electrical appliances in homes, their time and frequency of usage, including number and model of appliance. This sensitive personal data can accurately be deduced from energy metering data that is recorded by modern smart electricity meters provided by energy suppliers.
Once obtained through a smart meter, energy companies can employ AI to correlate this data with other publicly available information to provide private consumer insights such as age, demographics and household income. We illustrate how energy data, which is not necessarily considered to be sensitive data, can easily be transformed into sensitive and personal information due to function...