Discussion

ETHICS IN ARTIFICAL INTELLIGENCE

By

Christine Logan ,Kennedy Esochaghi ,Stone Koerner & Zachary Ellis

UNIVERITY OF MARYLAND GLOBAL CAMPUS

IFSM 304

5/8/2021

1

Software Application

Human Intellect

Machine Learning

Advanced Robotics

Virtual Agents/Worlds

Artificial Intelligence is the premise of implementing human intellect within computer software in order to solve complex tasks (Copeland, 2020). Artificial Intelligent technology works off a concept of machine learning which allows for the Artificial Intelligent technology to learn and adapt to new data without human interaction (Frankenfield, 2021). As Artifical Intelligence systems (robots, virtual agents, virtual worlds, and other advanced robotics) transition from being viewed simply as tools, and instead are now being viewed as entities and even team members, recognizing how these systems effect ethics is becoming increasingly important (Zhang et al., 2021).

2

How does AI Affect us?
Artificial Intelligence alters everyday lives

Artificial Intelligence affects human wellbeing, defense, and efficiency

Artificial Intelligence must be implemented in a manner that adheres to faith and empathy

Artificial Intelligence is now altering our everyday lives in ways that may be both beneficial and detrimental(Wrote may be also detrimental due to some slides talking about how AI can cause harm to jobs and etc.) to human wellbeing, defense, and efficiency. Transport infrastructure, service robotics, medical care, education, general protection and safety, and entertainment are all examples of this. However, to prevent such dystopian worlds from being a reality, these programs must be implemented in a manner that fosters faith and empathy, while still respecting civil and human rights (Dignum, 2018).
3

What ethical concerns does Artificial Intelligence impose?
Technology evolving creating smarter AI brings to questions whether machines can make smarter decisions over a human. This puts normal jobs and everyday tasks at risk.

If AI takes over everyday tasks does people’s privacy remain intact/or is it more exposed due to machines and their ability to be hacked maliciously

Ai put in a position to have access to people’s confidential interaction may cause the biggest ethical concern. If this device interacts with a hacker or person with malicious intent people’s data could be easily attacked and taken. Simply because the AI was not smart enough to know what decision to make.

AI puts normal jobs and everyday tasks at risk.

Machine learning allows for AI to learn and adapt without any human interaction.

From AI we get self-driving cars, job automation, gaming, and many more.

Self-driving cars themselves are an ethical issue. Can they be smart enough to avoid accidents and traffic?

If this device interacts with a hacker or person with malicious intent people’s data could be easily attacked and taken. Simply because the AI was not smart enough to know what decision to make.

Concerns have been raised of how technology could cause overreliance within industries and governments.
4

Does artificial intelligence pose a risk on job security for individuals?
Implementing AI into automated machines allows for robotics to operate longer and faster than a human.

Automation could make human interactions no longer needed in most industries.

AI can be looked at as an aide rather than a job taker. If humans can work with automated artificial intelligence the process in factories could become even more efficient that picking just one or the other.

Are jobs created when a machine takes a worker’s job. Who is going to fix this automated machine when it inevitably has a problem? Ethically machines are far from knowing when and how to make the right decision when posed with a problem and this can cause a big issue within a factory that is supposed to ruin smoothly to prevent down time and effectiveness.

If humans can work with automated artificial intelligence the process in factories could become even more efficient that picking just one or the other.

Are jobs created when a machine takes a worker’s job. Who is going to fix this automated machine when it inevitably has a problem?

this can cause a big issue within a factory that is supposed to ruin smoothly to prevent down time and effectiveness.

The big news would be that AI could make human interactions obsolete in some industries.
5

What does the future look like regarding artificial intelligence?
AI is said to become one day so intelligent that it could redesign itself and become weaponized when that is not what is wanted (Frankenfield, 2021)
Malicious hacking and invading personal privacy are a major concern when dealing with AI. A machine with AI is still a machine at the end of the day and can be attacked and controlled by parties that have intentions of using the device for a task it was not intended to complete.
At the current rate at which AI is developing 95
5 of human interaction will be powered by AI by 2025 and have the market reach a value of 118 billion dollars (about $360 per person in the US).

AI is extremely unlikely to become this intelligent however, it is a major concern for many when talking about the future of artificial intelligence inside of machine/robots.

A machine with AI is still a machine at the end of the day and can be attacked and controlled by parties that have intentions of using the device for a task it was not intended to complete.

Governments are already using AI for monitoring citizens so if this equipment were to become infected by a hacker it could result in massive amounts of confidential data being released.

The rapid growth of AI could result in potential problems for human employment, safety, and most importantly privacy.

6

How can biases in AI be eliminated?

Human vulnerabilities and prejudice must be removed from AI models.

Monitoring discrimination anomalies in AI software.

Anonymizing information is the most effective way for AI to maintain impartial results.

To eliminate inequalities, researchers should use statistics and data analysis to keep an eye out for anomalies. Artificial Intelligence discrimination is minimized at its most superficial level and avoided by assessing and verifying the representativeness of various samples of actual outcomes. Any Artificial Intelligence program that does not have prejudice control will eventually fail before applied to accurate output data to produce the utility that companies need, an Artificial Intelligence algorithm conditioned on a selected data collection. Anonymizing such information is the most effective way for Artificial Intelligence to maintain impartial results. Anonymizing may include suppressing names or other personally identifying details that could lead to prejudice.
7

How does society Keep a multifaceted intelligent machine under its control?
Use the Web’s globally accessible human intelligence to build optimization techniques where the algorithm is most likely to malfunction (Martin & Freeland, 2021).

Beat the Machine.

Experts in the field of AI need to remember the massive number of resources they have. Using the Web’s globally accessible human intelligence to help them think of possible scenarios before could allow them to stay ahead of any possible malfunctions.

To beat the machine makes it sound as if we are at war. Thinking of the movie War Games (1983). Much like the movie, humans identifying instances in which the machine’s projections are incorrect. When they succeed in detecting specific scenarios, the computer improves its ability to learn from them and cope with similar situations in the future. This technique aims to train the model to be more resilient by simply tossing in new and potentially unusual cases that are not present in the current data while testing the system.
8

How do we delineate the civilized handling of Artificial Intelligence?

Artificial Intelligence should be technically fit-for-purpose, as well as mentally and socially functional. Defining the humane handling of Artificial Intelligence, the system needs to have its legal rights. Humanity has a duty to our environment, and social structures and methods may use robots. We have a responsibility to safeguard and build them to defend themselves from abuse and be fundamentally aligned with civilization.
9

Legal Rights, for AI.

Humans’ ethical responsibility of the creation of AI.

How do we protect against Unintended consequences?

Regardless of who builds an AI system, they should be held accountable for any negative outcomes.

The Government should enforce that AI builders have a fail-safe system where all rogue systems can be taken offline

Imagine an AI system that is asked to eradicate cancer in the world. After a lot of computing, it spits out a formula that does, in fact, bring about the end of cancer – by killing everyone on the planet. The computer would have achieved its goal of “no more cancer” very efficiently, but not in the way humans intended it. With a computer system that functions on its own, how do we stop it?

10

How fast is ai growing?

The future of Artificial Intelligence is rapidly increasing and steadily being implemented within industries. Over the past four years Artificial Intelligence has increased 270 percent within industries (Biswal, 2021). It is estimated that by 2025, 95 percent of human interactions will be powered by Artificial Intelligence and that the Artificial Intelligence market will reach a revenue of 118 billion dollars (Biswal, 2021). The rapid development of future Artificial Intelligence technology has many concerned for their future employment, safety, and privacy.
11

Artificial Intelligence has seen a 270% increase within industries over the past 4 years

95% of human interactions powered by Artificial Intelligence by 2025

118-billion-dollar market for artificial intelligence

CONCLUSION

To define the humane handling of Artificial Intelligence, the system as a whole needs to have their legal rights. Humanity has a duty to our environment, and social structures and systems may use robots. We have a responsibility to safeguard and build them so that they can defend themselves from abuse and be fundamentally aligned with civilization. There is a slew of privileges they should be granted, such as the freedom to be guarded by our ethical and legal framework and the right to be built to be reliable. Therefore, Artificial Intelligence should be technically fit-for-purpose, as well as mentally and socially functional.
12

Responsible for safeguarding

Ethical and legal framework

Fit-for-purpose

Mentally and socially functional

ReFRENCES

Biswal, A. (2021, January 28). AI Applications: Top 10 Artificial Intelligence Applications. Simplilearn. Retrieved from https://www.simplilearn.com/tutorials/artificial-intelligence-tutorial/artificial-intelligence-applications

Copeland, B. (2020, August 11). Artificial Intelligence. Britannica. Retrieved from https://www.britannica.com/technology/artificial-intelligence/additional-info#history

Dignum, V. (2018). Ethics in artificial intelligence: introduction to the special issue. https://link.springer.com/article/10.1007/s10676-018-9450-z

Frankenfield, J. (2021, March 8). Artificial Intelligence (AI). Investopedia. Retrieved from https://www.investopedia.com/terms/a/artificial-intelligence-ai.asp

Zhang, Y., Wu, M., Tian, G. Y., Zhang, G., & Lu, J. (2021). Ethics and privacy of artificial intelligence: Understandings from bibliometrics. Knowledge-Based Systems, 106994. https://www.sciencedirect.com/science/article/abs/pii/S0950705121002574
Julia Bossmann.(2016,October 21) “Top 9 Ethical Issues in Artificial Intelligence.” World Economic Forum, www.weforum.org/agenda/2016/10/top-10-ethical-issues-in-artificial-intelligence/.