Ethics In Artificial intelligence Developers Need To Follow.

Artificial Intelligence

Artificial intelligence is a new boom in the technology industry. AI is no more a thing of the future. Now AI is around us. AI has become our present. In this post, we will talk about the importance and application of Artificial Intelligence and discuss how AI is changing our lives.

There are many notions and definitions of AI. Some say that it is an attempt to develop a simulation of the human brain, some say that it is an idea to teach machines how to learn, think and do stuff on its own. But for me, AI is just another complex and intelligent algorithm.

An algorithm that is able to do facial recognition, able to process natural language and able to process large structural and un-structural data.  So basically AI is an application of pure mathematics. It's just an algorithm with the neat mathematical calculation of numbers.

According to a recent study, it is found that around $16 trillion will be added between now and 2030 on the basis of AI. This is never before seen scale of economic impact. It just not impact the IT industry but every other industry out there.

Artificial intelligence means different to different industries. For a gaming industry, AI means better interaction with bots, interactive environment to its players. For a data-scientist, AI is a tool to extract meaning full information and classifying it for better understanding.

Natural language processing in AI is now becoming a way to enhance human and machine interaction. Alexa, Google Assistant, Siri are such application which can talk to humans. These technologies are enabling machines and humans to understand and interact with each other.

It is also creating new ways of doing business. Chatbots are such an example which provide an AI-based customer care environment. users can use chatbots to solve their general queries without any wait and hassle. It not only saves customers time but also saves the agents time to solve more complex queries.

In medical industries, chatbots are used for prescribing medicines based on the symptoms explained to them.

Computer vision applications such as facial recognition are now used as a security feature in your cell phones. It unlocks only when your face shows up. Law and enforcement industry is using computer vision applications to identify criminals in video feeds.

In the health industry, computer vision is used to identify cancer symptoms.

Even in entertainment, AI is playing an important role in giving recommendations on what to watch next based on our interest and watch lists. In shopping, AI is playing the role of a smart salesman which comes with better recommendations for other products we can buy!

AI is playing a role as a secret tool to detect fraudulent transactions. It can detect abnormal analogies and patterns.

In the health sector, AI is helping doctors come up with better diagnostics and results. With accurate and perfect timing of disease detection and diagnostic help.

Everywhere and all the time we are now surrounded by artificially intelligent systems that are making our lives secure and easy.

Ethics In Artificial Intelligence

Artificial Intelligence is a concept that is going to change many lives around the world. It is necessary to discuss the ethics and social considerations about the topic which is going to affect many lives around the world.

So, what is ethics? Why we need that? moreover, do we need ethics to work with science?

See, if we are talking about an ideal artificial intelligence concept, It is nothing but an alternate version of a human mind with capabilities to process the data, recognize and track patterns, understand the complex structure and notions way faster than a human mind can do. We may not have full access to the power of our brain because we still are wondering about its miracle, but AI, it is the sum total of the brain we understood till now.

So, it is very important to have a boundary on what is allowed and what is not allowed in artificial intelligence.

Let's take an example, someone invented the notion of a knife. You can use that knife for many purposes say to cut fruit, vegetables or for cooking purposes or say creating hand-crafts, but at the same time, it can be used for murder, kill violence.

So, ethics is not a technological problem. It is a human problem, It is up to us to how to deal with such advanced technologies.

Ethical Concerns for AI Developers

AI is extremely powerful and can be used in ways that can negatively affect society. To develop and use the AI system responsibly, AI developers must consider the ethical issues inherent with AI.

Four Pillars Of Ethics

Transparency
Accountability
Privacy
Lack Of Bias

Bias

One of the key notions to keep in mind is the bias present in an AI model.  An AI model should be bias-free, It should be free of any bias say, gender, racial or social.

One of the best examples comes up when an image recognition system shows the kitchen, laundry and households item with women and sports, travel, coaching, adventures with men. This happens due to poor selection of training data that have the bias contents while training the model.

Bais is the most important thing to keep in the mind for an AI developer. Recently in an image recognition system, it is found that it can recognize people with a lighter tone more efficiently than skin with a darker tone. Such models can create a disaster while applying in the real world scenario and be a case of racism.

AI developers must guard their model against such biases by the uses of effective and bias-free training data. They must ensure that the AI system is free of biases before landing on the world of humans.

Developing Trust With Transparency

Another Ethical issue that has arisen is the person entitled to know when they are speaking to a bot and when they are speaking to a human being. Now a day bot conversation has developed much like a human conversation. Due to a lack of trust in the AI system, If a person recognizes that he is talking to a bot rather than a human being, it can be unsettling and can make the transaction feel unequal. Transparency is the key, people should be aware when they are interacting with bots and can manage their expectations accordingly.


We talked about Transparency and bias. let's talk about Accountability and privacy.

Accountability

AI developers should create an AI system with algorithmic accountability so that any unexpected behavior can be traced and undone.

Privacy

Personal Information Should Always Be Protected.

Social Concerns

Artificial intelligence is not a future anymore. It is around us. Slowly we are implementing it in our day to day lives.

The changes from the AI system are inevitable. It will impact in a large volume. Some people are saying that artificial intelligence is going to be the last and grand invention of human beings.

I, however, disagree with "the last invention" cause we don't know what is cooking under the brain. I mean till now we did not know our brain 100%, I mean yes we have made an advancement in biological dimensions, but there are still other dimensions of the brain we don't know yet.

So to say that AI is the last and ultimate invention should be wrong. To say that AI will replace humans, it is also not possible as ultimately AI is just a piece of a program, it is just a notion of our mind.

But yes, AI is one of the grand inventions, and we must talk about how to use it properly. It is not the machine who will become bad, It will be the human intention that can go bad.

There are some social concerns out there, For example, say, In a self-driving car what if it needs to choose between the incidents or accidents. If it has to pick between running into a sign and hurt the passenger in order to save the pedestrian or running over the footpath and hurt the pedestrian, but potentially save the passenger. How can we deal with such problems?

See, the above situation is an emotional problem if occurred to a human being. If a human was driving the car and the situation comes, it depends upon the emotion and good wills of the driver to choose what to happen. But for a machine, what is more important? to save the passenger or to save the pedestrian?

You can think of these social concerns with a very interesting story.

If we go some 100 years back, we have a horse carriage. suppose a person was driving that horse carriage. A pedestrian passing by humbly whistling and enjoying his day, the horse did not like the smell or the sound and got irritated and the horse trampled the pedestrian.

Now, whom you will blame? The rider has no control over the horse emotion, the pedestrian was just passing by and we can't punish a horse. it will not make any sense.

Similarly, social concerns around the AI is the same. We have to take the steps carefully and within the boundary of AI ethics, we talked above in detail.