Study claims 40% of domestic work could be done by robots within a decade

According to a survey of 65 AI experts in the UK and Japan, robots could perform around 39% of household tasks within the next decade, with grocery shopping being the most likely to be automated. However, caring for the young and old is less likely to be affected. The increased use of AI in homes could also result in privacy concerns. One of the report's authors, Ekaterina Hertog, suggested that society needs to have a public discussion about privacy in the age of smart technology, as per a report from the Guardian. Despite this, she believes that greater automation could lead to improved gender equality by reducing the burden of unpaid work, which is mainly borne by women. In the UK, women do over twice as much unpaid work as men, while in Japan, men do less than one-fifth of the unpaid work done by women.

According to the study published in the journal Plos One, only 28% of care work such as caring for the elderly or teaching, is expected to be automated, while 60% of time spent on grocery shopping may be reduced. Despite predictions made for several decades, the reality of domestic robots being able to perform tasks such as putting out the bins and picking up toys remains elusive. Hertog compared the optimism surrounding domestic robots to that of self-driving cars, which has yet to navigate the unpredictable environment of the streets. In the same way, homes pose similar challenges for robots to function effectively.

What does Moravec's paradox say?

Moravec's paradox is a concept in artificial intelligence that was first introduced by Hans Moravec, an AI researcher, in the 1980s. The paradox states that tasks that are easy for humans to perform, such as recognizing faces or walking, are difficult for machines to do, while tasks that are difficult for humans, such as complex mathematical calculations or data processing, are relatively easy for machines to do.

Moravec's paradox arises from the fact that humans and machines approach problem-solving in different ways. Humans have evolved over millions of years to be able to perform tasks that are essential for survival, such as recognizing faces, walking, and interacting with the environment. As a result, these tasks are largely carried out by the subconscious mind, which is a very efficient and specialised system that processes information in parallel.

Machines, on the other hand, are designed to perform tasks that require logical reasoning and mathematical calculations. These tasks can be easily programmed into machines and can be carried out with great speed and accuracy. However, machines struggle with tasks that require context, common sense, and situational awareness.

Moravec's paradox has significant implications for the development of AI. It suggests that the tasks that are currently seen as difficult for machines, such as understanding natural language, may remain difficult for a long time to come, while tasks that are currently seen as easy for machines, such as data processing and mathematical calculations, may become increasingly automated.