When somebody mentions artificial intelligence, people think of science fiction robots or a dystopian future. While pop culture and the media depict AI in a sceptical fashion, these ideas of AI are often exaggerated for film and TV.

What is Artificial Intelligence?

The term ‘Artificial Intelligence’ was first coined in 1955 by renowned American Computer & Cognitive Scientist John McCarthy. It appeared in a proposal submitted by McCarthy and fellow academics, who defined AI as ‘the science and engineering of making machines’. The concept had already been explored in science, mathematics, philosophy, and literature (e.g., the ‘heartless’ Tinman character in The Wizard of Oz). British polymath Alan Turing notably investigated the mathematical prospect of AI in his 1950 paper ‘Computing Machinery and Intelligence’. He proposed that a human’s ability to use available information combined with reason in order for problem-solving and decision-making could be replicated by machines.

AI’s Upsides

Most of us encounter AI every day. Anything is AI if it involves a program doing an action or task that would usually rely on human intelligence. For example, in India, Samsung recently brought out a line of AI-enabled bi-lingual washing machines using the company’s EcoBubble and QuickDrive time and power-saving technology (Source: Mint). The new models learn and remember laundry habits, suggest the most frequently used wash cycle, and recommend better wash cycles for the user. What’s not advantageous about that?

With AI, machines can work 24/7 and don’t need breaks or a work-life balance. This is particularly helpful in jobs like customer service, where AI ‘chatbots’ can be on standby to answer customer queries across different time zones.

AI is also having a positive impact on medicine in many different ways. For example, it is being used to detect and diagnose diseases earlier on and more accurately, improving doctors’ prognoses and disease management. Robots are also used for a multitude of medical tasks – from surgical procedures to physiotherapy (Source: PWC). AI was also widely used during the COVID-19 pandemic to understand, track, and prevent the spread of the virus. It helped determine certain risk groups, aided the creation of contact tracing apps, and reduced hospital staff’s workloads with testing methods that minimised human labour and contact (Source BBVA).

AI’s Downsides

AI applications such as Siri and Alexa are part of people’s daily lives globally, and we subconsciously use them to complete many tasks. However, such applications often come with a ‘data gap’, where AI mirrors certain human biases. Award-winning author Caroline Criado Perez writes a lot about this in her book Invisible Women where she explains that a ‘male default’ is often entrenched in such technology. She uses the smartphone market as an example, noting that, while the average smartphone ‘fits comfortably’ in the average man’s hand, it’s barely bigger than the average woman’s hand. Seeing as evidence shows women are more likely to be iPhone owners, this is ironic. Perez also discusses a ‘female-friendly’ Chinese smartphone called the ‘Keecoo K’ - a device designed for smaller hands, but at the cost of inferior processing power and in-built airbrushing… Additionally, the female identity of Amazon’s Alexa and other smart-home technologies can perpetuate antiquated views of women in the domestic sphere.

The use of AI in policing has also had worrying consequences. In the US, police have used facial recognition software to aid investigative processes like identifying a suspect with an arrest warrant. However, in 2018, MIT published a study revealing software error rates ranging from 0.8% for light-skinned men to 34.7% for dark-skinned women when using this type of software. An example error was when Nijeer Parks was arrested in 2019 for shoplifting and attempting to hit a police officer with a car, despite the crime taking place 30 miles from his home and owning neither a car nor a license. Parks was also held for ten days and denied due process based on an algorithmic recommendation (Source: Teen Vogue).

Moreover, from remembering our laundry cycles to quite literally saving lives, there’s no denying the benefits of AI. However, it’s crucial that we are aware of ‘algorithmic injustice’ and that we are implementing the correct strategies to challenge biased AI.

 

Published By
Ruth Mullan