Concerns and Ethics of Artificial

Intelligence
Intro to AI and Data Science
NGN 112 – Fall 2024

Ammar Hasan
Department of Electrical Engineering
College of Engineering

American University of Sharjah

Prepared by Dr. Salam Dhou, CSE

Last Updated on: 22nd of August 2024

 Table of Contents
2

Concerns of Artificial Intelligence (AI)

Ethics in AI

How to establish AI Ethics

 Primary Concerns of AI
3

 There are a number of issues that are at the

forefront of ethical conversations surrounding AI
technologies. Some of these include*:
1. Technological singularity
2. AI impact on jobs
3. Privacy
4. Bias and discrimination
5. Accountability
6. Environmental Impact

*https://www.ibm.com/topics/machine-learning
 Concern
1
Technological Singularity
4

 Technological singularity
 While this topic garners a lot of public attention, many
researchers are not concerned with the idea of AI
surpassing human intelligence in the near or immediate
future.

Figure from Colvin Consulting Group

 Concern
Technological singularity
1
(cont.)
5

 Technological singularity, Cont.

 Despite the fact that Strong AI and superintelligence is not
imminent in society, the idea of it raises some interesting
questions as we consider the use of autonomous systems,
like self-driving cars.
 It’sunrealistic to think that a driverless car would never get into a
car accident, but who is responsible and liable under those
circumstances?
 Should we still pursue autonomous vehicles, or do we limit the
integration of this technology to create only semi-autonomous
vehicles which promote safety among drivers?
 The jury is still out on this, but these are the types of ethical
debates that are occurring as new, innovative AI technology
develops.
 Concern
The Three Laws of Robotics
1
(or AI)
6

 In his 1950 book, “I, Robot”, Isaac Asimov

predicted the rise of robotics (and AI), and
outlined “The Three Laws of Robotics”:
1. A robot may not injure a human being or,
through inaction, allow a human being to
come to harm;
2. A robot must obey the orders given to it
by human beings except where such
orders would conflict with the First Law;
3. A robot must protect its own existence as
long as such protection does not conflict
with the First or Second Law;
 Concern
2
AI Impact on Jobs
7

 AI impact on jobs
 While a lot of public perception around artificial intelligence
centers around job loss, this concern should be probably
reframed.
 With every new technology, we see that the market demand for
specific job roles shift.
 For example, when we look at the automotive industry, many
manufacturers, like GM, are shifting to focus on electric vehicle
production to align with green initiatives. The energy industry
isn’t going away, but the source of energy is shifting from a fuel
economy to an electric one.
 Concern
2
AI Impact on Jobs (cont.)
8

 AI impact on jobs, Cont.

 Artificial intelligence should be viewed in a similar manner,
where artificial intelligence will shift the demand of jobs to other
areas.
 There will need to be individuals to help manage these systems
as data grows and changes every day.
 The important aspect of artificial intelligence and its effect on
the job market will be helping individuals transition to these new
areas of market demand.
 Concern
3
Privacy
9

 Privacy
 Privacy tends to be discussed in the context of data privacy, data
protection and data security, and these concerns have allowed
policymakers to make more strides here in recent years.
 For example, in 2016, GDPR legislation was created to protect the
personal data of people in the European Union and European Economic
Area, giving individuals more control of their data.
 In the United States, individual states are developing policies, such as the
California Consumer Privacy Act (CCPA), which require businesses to
inform consumers about the collection of their data. This recent legislation
has forced companies to rethink how they store and use personally
identifiable data (PII).
 In UAE, TDRA is the government entity responsible with data protection
and security.
 As a result, investments within security have become an increasing
priority for businesses as they seek to eliminate any vulnerabilities
and opportunities for surveillance, hacking, and cyberattacks.
 Concern
4
Bias and Discrimination
10

 Bias and discrimination

 Bias and discrimination can be found in a number of
applications from facial recognition software to social media
algorithms.
 Machine learning algorithms need big datasets to be used for
training. The more data you put in, the more accuracy you get
out. Since minorities are by definition less well-represented in
the population than the majority, a lack of data may explain
much of the “bias” machine learning systems.
 Instances of bias and discrimination across a number of
intelligent systems have raised many ethical questions
regarding the use of artificial intelligence.
 How can we safeguard against bias and discrimination when
the training data itself can lend itself to bias? Discussed in next
slides (Ethics in AI)
 Concern
5
Accountability
11

 Accountability (Who is responsible when AI goes

wrong?)
 Who should be held accountable when AI misbehaves? The users,
the creators, the vendors?
 The lack of standardized guidance around AI governance, and the
complexity of deep learning and machine learning models, has
made it difficult for experts to decide who to blame when an AI
system goes wrong.
 Since there isn’t significant legislation to regulate AI practices,
there is no real enforcement mechanism to ensure that ethical AI is
practiced.
 Ethical frameworks have recently emerged as part of a
collaboration between ethicists and researchers to govern the
construction and distribution of AI models within society.
 Concern
Managing AI Environmental
6
Impact
12

 Managing AI environmental impact

 AI models are getting larger every day, with recent models
being as large as over a trillion parameters each.
 These large models consume a significant amount of
energy to train - making AI a large resource consumer.
 Researchers are developing techniques for energy efficient
AIs that balance performance and energy efficiency.
13

Ethics in AI
 Ethics in AI
14

Definition
 AI Ethics is the term given to a broad collection of

considerations for responsible AI that combines safety,

security, human concerns and environmental
considerations. *

 Some areas of AI Ethics include*,**:

 Avoiding AI Bias
 AI and Privacy
 Responsible use of energy and resources
 Avoiding AI mistakes

*Nisha Talagala, Forbes:

https://www.forbes.com/sites/nishatalagala/2022/05/31/ai-ethics-what-it-is-
and-why-it-matters/?sh=6f492aa33537
 Avoiding AI Bias
15

 Avoiding AI Bias
 Human beings come with all sorts of cognitive biases, such as recency and
confirmation bias, and those inherent biases are exhibited in our behaviors
and subsequently, our data.
 Since data is the foundation for all machine learning algorithms, it’s
important for us to structure experiments and algorithms with this in mind
as artificial intelligence has the potential to amplify and scale these human
biases at an unprecedented rate.
 Since AI models learn from data, poorly constructed AIs can demonstrate
bias against poorly represented subsets of the data.
 In particular, AI models that are not well-trained can demonstrate bias
against minorities and underrepresented groups. Famous cases of bias,
such as in hiring tools and in chatbots, have embarrassed famous
corporate brands and created legal risk.
 AI and Privacy
16

 AI and Privacy
 AI relies on information to learn from. A significant fraction of
this information comes from users.
 Not all users are aware of what information is being gathered
about them and how it is being used to make decisions that
affect them.
 Even today, everything from internet searches to online
purchases to social media comments can be used to track,
identify, and personalize experiences for users.
 While this can be positive (such as an AI recommending a
product a user may want), it can also result in unexpected
bias (such as some offers being provided to some consumers
and not others).
 Avoiding AI Mistakes
17

 Avoiding AI mistakes
 Poorly constructed AIs can make mistakes which can lead
to anything from loss of revenue to death.
 Adequate testing is required to ensure that AIs do not pose
a risk to humans or to their environments.
 How to Establish AI Ethics
18

 Teams have started to assemble frameworks

and concept to address some of the current
ethical concerns.
 There is some consensus around incorporating:

 Governance
 Explainability
 How to Establish AI Ethics:
Governance
19

 Governance:
 Companies can leverage their existing organizational
structure to help manage ethical AI.
 If a company is collecting data, it has likely already
established a governance system to facilitate data
standardization and quality assurance.
 Internal regulatory and legal teams are likely already
partnering with governance teams to ensure compliance
with government entities, and so expanding the scope of
this team to include ethical AI is a natural extension of its
current priorities.
 This team can also steward organizational awareness and
incentivize stakeholders to act in accordance with
company values and ethical standards.
 How to Establish AI Ethics:
Explainability
20

 Explainability:
 Machine learning models, particularly deep learning models,
are frequently called “black box models” as it’s usually
unclear how a model is arriving at a given decision.
 Explainability seeks to eliminate this ambiguity around
model assembly and model outputs by generating a “human
understandable explanation that expresses the rationale of
the machine”.
 This type of transparency is important for building trust with
AI systems to ensure that individuals understand why a
model is arriving to a given decision point.
 If we can better understand the why, we will be better
equipped to avoid AI risks, such as bias and discrimination.
 How to Establish AI Ethics: Closing
Words
21

 Achieving ethical AI will undoubtedly be

important to its success.
 However, it’s important to note that it has

tremendous potential to impact society for

good.
 This is started to be evident in its integration

into areas of healthcare, such as radiology.

 This conversation around AI ethics is to ensure

that in our attempt to harness this technology

for good, we appropriately assess its potential
for harm within its design.
 Learning Outcomes
22

Upon completion of the course, students will be

able to:
1. Identify the importance of AI and Data

Science for society

2. Perform data loading, preprocessing,
summarization and visualization
3. Apply machine learning methods to solve basic
regression and classification problems
4. Apply artificial neural networks to solve simple
engineering problems
5. Implement basic data science and machine
learning tasks using programming tools