February 06, 2023 Ethics and AI: What You Need to Know About Including Ethics in AI

By Sam Ancer

Artificial Intelligence (AI) has the power to change many different industries, like self-driving cars, healthcare, law enforcement, and even the court of law. 

Before we start using Artificial Intelligence, we need to think about what could happen. For example, if a self-driving car crashes into a pedestrian, who should be responsible? What if the AI is not fair and hurts certain groups of people? 

What will happen to people's jobs if machines take over their tasks? And what if Artificial Intelligence is used to copy someone else's creative work without giving them credit? 

These are all important questions that we need to answer before using AI. It's important for our society to be prepared and make sure that we use Artificial Intelligence in an ethical way.

Ethics and AI Self-Driving Cars

Artificial intelligence is being brought into the transportation sector, specifically self-driving cars. Though these cars have the potential to improve safety and efficiency, the question of responsibility arises. 

If an Artificial Intelligence self-driving car causes an accident or breaks traffic laws, who is accountable - the car manufacturer, AI developer, or the individual in the vehicle?

There are also questions about what a self-driving vehicle should do in a given situation. These include ideas around who should be favoured in an accident, pedestrians or the vehicle's passengers. 

There is also the question of if someone is breaking the law, should that be a factor in the decision making of the Artificial Intelligence operated vehicle. Is it just a question of saving the most amount of lives? 

These are the ethical concerns brought up by Moral Machine, a platform that presents moral dilemmas faced by autonomous vehicles and crowdsources opinions on ethical decision-making in AI-powered self-driving cars.

In a lot of ways ethical concerns around AI self-driving cars are the easiest problems to understand, because their impact has such visceral consequences. 

One thing that is a big concern is Artificial Intelligence and bias.

Ethics and AI Bias

Bias in AI means that the computer programs that power Artificial Intelligence systems have mistakes or unfair ideas built into them. This might not be on purpose, but may just be because of the datasets used in the learning models to develop AI. 

The impact is the same, where people of colour, women, and other underrepresented groups are affected unfairly. For instance, an AI can help determine your eligibility for a home loan. If it is fed information incorrectly, it could determine that certain population groups should not qualify for a home loan. 

This could lead people to be unable to access housing and could result in massive financial insecurity for swathes of the global population.

Beyond that there are concerns with how we will use Artificial Intelligence in medicine, law enforcement, and even the judicial system.

If we do not correct institutional biases there could be a number of unforeseen consequences due to inherent biases that were not picked up by the developer of the AI.

The goal of AI is to help our society advance and to optimise processes. Neither of these things are possible if Artificial Intelligence is being used to continue our biases and other flaws in our thinking.

While bias in Artificial Intelligence is concerning and needs to be dealt with, there is also the issue of AI and automation.

Ethics and AI Automation

One of the most exciting ideas around Artificial Intelligence is the idea of automating certain tasks and processes. There are concerns that this process of automation will reduce jobs and lead to poverty or worse for vulnerable members of society.

Champions of automation believe that by automating parts of peoples jobs this will rather make their jobs easier than completely replace people.
One concern with this line of thinking is that, if AI is able to make someone ten times more productive, would the company just remove the 9 other employees and give their work to the single employee and their Artificial Intelligence tools?

While it's not in our immediate future to be replaced by robots, things like BuzzFeed replacing 180 employees' roles by empowering their other staff with ChatGPT does leave us with some cause for concern.

It is crucial that our society gears itself up to take care of the people whose careers are soon to be impacted by the power of Artificial Intelligence. 

Another point of concern for ethics and AI is the idea of Artificial Intelligence plagiarism.

Ethics and AI Plagiarism

Everything a machine learning Artificial Intelligence does is something it has “learned” from a data set. 

Essentially an AI will be given a number of parameters and then will be fed information that does or does not contain those parameters. It will then receive instructions and use the data it has, along with its parameters, to perform those instructions in whatever way it can.

When Artificial Intelligence is used to create art, it will examine art pieces that hold specific tags, for instance an art piece will be tagged as surreal, and then the qualities the AI can identify with the artpiece, it will then associate as surreal. The larger the data set an Artificial Intelligence has, the more precise it can be.

An issue comes in the fact that pieces of creative work, whether they be written, painted, or recorded, are being used to create art pieces without appropriate attribution to the artist who influenced the generated piece.

This argument does not hold up to the idea of homage or influence in art. It is common for an artist to be inspired by a number of pieces without having to directly reference or pay for the other artist's work.

If the people who are concerned with AI plagiarism get what they want, it could mean the end of creative expression as people would be forced to develop wholly original works without reference to any other works, aside from those that exist in the public domain or creative commons.

Regardless it does seem unfair that a programmer can use someone else’s art to make money and not even reference that artwork.

Ethics and AI: What You Need to Know About Including Ethics in AI

Artificial intelligence has the power to change many industries, but before we fully use it, we need to think about the ethical side of things. 

We need to answer important questions about things like self-driving cars, bias in AI, Artificial Intelligence taking over jobs, and AI copying others' work. 

This article has not even touched on concerns like AI and Data gathering, the government's role in regulating AI, and who should be held accountable in moral concerns around AI. 

Whether we like it or not, AI is going to have a massive impact on our daily lives, mostly for the better, but it's vital that we ask questions around Artificial Intelligence and ethics right now so that we can solve any issues before they begin