AN ASSIGNMENT ON:

DECODING THE ETHICAL

AI: EUROPEAN UNION
ACT

Name: V. Sadvik Kumar

Course: CSE AIML
Roll No: 23C71A6601
Date of Submission: 14-04-2025
CONTENTS

Introduction
Historical context and legislative Framework
Objectives of the AI Act
Structure of the AI Act
Risk-Based Categorization
High-Risk AI Systems – A Deep Dive
Ethical Foundations of the Act
Compliance Mechanisms
Challenges and Criticisms
Global Influence and Geopolitical Implications
Role of Stakeholders
Case Studies and Hypothetical Scenarios
The Future of Ethical AI
Algorithm: EU Ethical AI Act Compliance
Process
Advantages and Disadvantages
Applications
Conclusion
References
INTRODUCTION

The rapid advancement of Artificial Intelligence (AI) has

transformed multiple sectors, from healthcare and education
to finance and entertainment. While AI offers great potential
for innovation, it also presents significant risks. These risks
include the potential for bias, discrimination, and violations
of privacy, making regulation essential. In response, the
European Union (EU) introduced the AI Act, the first
comprehensive legal framework designed to regulate AI
technologies based on their perceived risk level.
The AI Act is critical in providing clear guidelines for the
development, deployment, and use of AI systems. It aims to
ensure that AI technologies adhere to ethical principles, such
as fairness, transparency, and accountability. This act is not
just a response to the risks posed by AI but also an effort to
provide legal certainty for businesses and developers within
the EU. By setting clear standards, it also fosters innovation
by creating a safe environment for businesses to develop AI
technologies without fear of unexpected regulatory hurdles.
As AI becomes an integral part of daily life, the European
Commission has taken the lead in creating regulations that
protect both individuals and society while still enabling
technological advancements. The AI Act reflects the EU’s
commitment to human rights and democratic values, making
it an essential milestone in the global discourse on AI
governance.

HISTORICAL CONTEXT AND LEGISLATIVE

FRAMEWORK
Before the introduction of the AI Act, the European Union
had already established various regulations concerning digital
and data privacy, with the General Data Protection Regulation
(GDPR) being the most prominent. The GDPR, which went
into effect in 2018, focuses on protecting individuals’
personal data and privacy rights in the digital world. While
GDPR set a significant precedent, it did not directly address
AI's unique challenges.
Building on GDPR's success, the EU sought to expand its
regulatory framework to specifically tackle the evolving
challenges presented by AI. The AI Act is an essential
extension of the EU's existing regulatory structure and comes
in conjunction with other digital governance laws, such as the
Digital Services Act (DSA) and the Digital Markets Act (DMA).
These regulations collectively shape the future of digital
governance in Europe, with a shared aim of creating a fair,
transparent, and secure digital environment.
The development of the AI Act also reflects the EU’s ongoing
efforts to regulate emerging technologies in a way that
safeguards the interests of its citizens. It draws inspiration
from the GDPR's framework of protecting individual rights
while embracing new technologies, offering a regulatory
model that emphasizes both safety and innovation.

OBJECTIVES OF THE AI ACT

The primary objective of the AI Act is to ensure that AI
systems are safe, ethical, and aligned with human rights. The
act uses a risk-based approach, classifying AI applications into
categories based on the level of risk they pose to individuals
or society. This method helps ensure that high-risk AI systems
are subject to strict regulatory oversight, while low-risk
systems can enjoy more flexibility.

A key goal of the AI Act is to safeguard fundamental rights,

such as privacy and dignity, and to protect individuals from
potential harms like discrimination, bias, and breaches of
confidentiality. It also aims to create a level playing field
within the EU by harmonizing AI laws across member states.
This legal certainty is critical for businesses, as it helps them
understand the specific requirements they must meet to
bring AI systems to market.
The act further seeks to foster innovation by promoting
trustworthy AI. By establishing ethical standards, the AI Act
aims to provide clear guidelines for the development of AI
technologies that do not compromise safety or fairness. In
this way, the Act attempts to create a balance between
enabling technological growth and safeguarding public
welfare.

STRUCTURE OF THE AI ACT

The AI Act is structured to regulate different aspects of AI
systems, categorizing them based on the potential risks they
pose. A key component of the Act is the definition of AI,
which is broad and encompasses various techniques,
including machine learning, expert systems, and logic-based
systems. These definitions are crucial for determining which
technologies fall under the scope of the Act.
The Act primarily targets AI providers, those who develop and
deploy AI technologies, and users, who integrate these
systems into their operations. Distributors and importers of
AI systems within the EU are also covered by the regulation,
ensuring that the entire supply chain is subject to oversight.
One of the most important aspects of the AI Act is its
categorization of AI systems according to their risk. The Act
aims to ensure that high-risk AI systems—those that have the
potential to significantly impact individuals’ lives—undergo
stringent regulatory measures. Meanwhile, lower-risk
systems are subject to less oversight, with the goal of
promoting innovation while ensuring ethical deployment.

RISK-BASED CATEGORIZATION
The AI Act introduces a risk-based categorization system that
is foundational to its approach to regulation. AI systems are
divided into four categories based on the level of risk they
pose:
 Unacceptable Risk: This category includes AI systems
that pose significant threats to public safety and
individual rights. Examples include AI used for social
scoring, as seen in some countries, and real-time
biometric surveillance (e.g., facial recognition in public
spaces). These applications are prohibited by the AI Act
due to their potential for harm and exploitation.

 High-Risk: High-risk AI systems are those that can

significantly affect people's lives, such as in areas like
healthcare, law enforcement, and finance. For these
systems, the Act mandates specific safety measures,
including human oversight, risk assessments, and
technical documentation. Examples include medical
 diagnostic tools and AI systems used in recruitment or
credit scoring.

 Limited Risk: AI systems with limited risk, such as

chatbots, must meet transparency requirements. For
example, chatbots must disclose that they are AI-driven
to ensure transparency in interactions with users.

 Minimal Risk: These systems are deemed safe and

present little to no risk. AI used for entertainment or
basic tasks (like spam filters or video game AI) falls into
this category. These systems are minimally regulated,
allowing for a flexible environment for innovation.
This tiered approach ensures that the most potentially
harmful AI systems are tightly regulated, while lower-risk
systems benefit from less stringent oversight.

HIGH-RISK AI SYSTEMS: A DEEP DIVE

High-risk AI systems are subject to the strictest regulations
under the AI Act. These systems can have significant
implications on human well-being, including health,
employment, education, and justice. Some examples of
high-risk systems include AI used in medical devices,
autonomous vehicles, and criminal justice systems.
High-risk systems must undergo a conformity assessment
before being deployed, ensuring they meet the necessary
safety and ethical standards. The AI Act requires that high-
 risk systems maintain technical documentation that
outlines their design, development process, and potential
risks, ensuring full transparency.
Moreover, human oversight
is required in many high-risk
AI applications. For example,
AI used in hiring decisions
must allow human
intervention to prevent
biased decisions. This is
especially important to
ensure that AI systems
operate in ways that align with fundamental rights, such as
non-discrimination.

By mandating these rigorous standards, the AI Act ensures

that high-risk systems are built with safety, accountability,
and transparency at the forefront.

ETHICAL FOUNDAIONS OF THE AI ACT

The ethical foundation of the AI Act is central to its design.
The EU is focused on ensuring that AI systems do not
undermine fundamental human values such as privacy,
freedom of expression, and equal treatment. As AI
technology evolves, the potential for misuse increases, which
makes it crucial to embed ethical principles in the
development and use of AI.
One of the key ethical principles emphasized in the Act is
transparency. AI systems must provide clear and
understandable information about how they function and
make decisions. This ensures that users are fully aware of the
algorithms that shape their interactions and outcomes,
whether in hiring, medical diagnoses, or financial services.
Another principle is accountability. The AI Act establishes that
the developers, deployers, and users of AI systems must take
responsibility for the impact of these systems. If an AI system
causes harm, the responsible parties must be held liable. This
principle reinforces the need for robust documentation and
testing before deployment.
The Act also aims to promote non-discrimination by ensuring
that AI systems are designed to avoid biases based on race,
gender, age, or other personal attributes. For instance, AI
used in recruitment must be scrutinized to prevent
discrimination against candidates from certain demographic
groups. The goal is to ensure that AI contributes to fairness,
not inequality.
Lastly, the human-centric approach is a fundamental part of
the Act. It asserts that AI should always serve human
interests and be aligned with democratic values. AI should
never be used to undermine human dignity or rights. By
maintaining human oversight and fostering human-in-the-
loop approaches, the Act guarantees that AI systems cannot
make final, irreversible decisions without human
intervention.
These ethical principles are crucial to creating a framework
where AI can thrive without compromising the values that
underpin democratic societies. The European Commission’s
emphasis on ethics places human rights at the heart of AI
regulation, distinguishing the EU’s approach from other
regions that may prioritize technology development over
societal impact.

COMPLIANCE MECHANISMS
The compliance mechanisms of the AI Act are designed to
ensure that AI systems meet regulatory standards before they
are deployed in the market. These mechanisms focus on
monitoring the entire lifecycle of AI systems, from
development through to usage.

One of the primary compliance tools in the Act is CE marking.

In the EU, products that comply with regulatory standards
must carry the CE mark, which signifies conformity with the
EU’s safety, health, and environmental protection standards.
For AI systems, this means they must undergo a conformity
assessment to ensure they meet the requirements outlined in
the AI Act, particularly for high-risk systems. The assessment
must cover a variety of factors, including safety, fairness,
transparency, and potential impacts on human rights.
To oversee the implementation and adherence to the AI Act,
the Act establishes national supervisory authorities in each
EU member state. These authorities are responsible for
monitoring AI systems, conducting audits, and ensuring that
AI providers and users comply with the law. They also handle
complaints and report non-compliance, acting as the primary
enforcement bodies within their respective countries.

Non-compliance with the Act can result in severe penalties,

as the Act includes provisions for imposing fines on
companies that fail to adhere to its regulations. These fines
can be substantial—up to €30 million or 6% of global annual
turnover, whichever is greater. These penalties act as a
deterrent, ensuring that companies take the compliance
process seriously and implement AI systems responsibly.
Through these mechanisms, the AI Act ensures that AI
systems are not only safe and ethical but that there are
concrete enforcement measures to hold developers and users
accountable. This structured compliance process is key to
ensuring the Act’s effectiveness in regulating the evolving
landscape of AI.

CHALLENGES AND CRITICISMS

While the AI Act is groundbreaking in its approach to
regulating AI, it is not without its challenges and criticisms.
One of the main criticisms is the concern that the Act could
lead to over-regulation, particularly for smaller companies
and startups. The stringent requirements for high-risk AI
systems—such as conducting conformity assessments,
providing detailed technical documentation, and ensuring
human oversight—can be costly and resource-intensive.
Smaller companies, especially startups, may struggle to
comply with these rules, potentially stifling innovation or
pushing businesses to relocate to jurisdictions with less
regulation.
Another criticism is the vague definitions present in the Act.
Some terms, such as “high-risk” and “unacceptable risk,” are
open to interpretation, which could lead to legal uncertainty
and inconsistency in enforcement. As AI technology evolves
rapidly, these definitions may need to be adjusted to keep
pace with emerging innovations like generative AI or
foundation models (large, pre-trained AI systems that can be
adapted to a variety of applications). If the Act does not
address these newer forms of AI, there could be regulatory
gaps that leave these technologies unregulated or under-
regulated.

Furthermore, some argue that the Act’s focus on human

oversight may not be feasible in all AI applications. For
instance, AI used in complex, real-time decision-making
systems (such as autonomous vehicles or advanced medical
AI) may require immediate responses that humans cannot
always provide. Balancing the need for human oversight with
the speed and efficiency required by certain AI applications
presents a challenge for regulators and developers alike.

Despite these challenges, the AI Act represents an ambitious

attempt to ensure the ethical development of AI. As AI
continues to advance, the Act may evolve and adapt to
address emerging concerns, but its current framework
provides a much-needed foundation for regulating AI in the
EU.

GLOBAL INFLUENCE AND GEOPOLITICAL

IMPLICATIONS
The AI Act has far-reaching global implications, especially for
countries outside the EU. The Brussels Effect refers to the
phenomenon where EU regulations set global standards,
influencing companies and governments worldwide. The AI
Act is likely to have a similar impact, as many global
companies that wish to operate in the EU will need to adhere
to its standards, even if they are based outside the EU.
As the first comprehensive regulatory framework for AI, the
AI Act could become a blueprint for other regions. Countries
like the United States, China, and Canada have been exploring
their own AI regulations, but they have taken different
approaches. For instance, the US has tended to favor industry
self-regulation, while China focuses on state control over AI
development, especially in surveillance and military
applications.

In contrast, the EU's approach to AI regulation is grounded in

human rights and ethical principles, setting it apart from the
more market-driven or state-driven models. If other countries
adopt similar regulations, it could create a more global
regulatory landscape for AI, ensuring that AI technologies are
developed and deployed responsibly around the world. The
Act also signals that the EU is positioning itself as a global
leader in tech governance, pushing for a regulatory
framework that prioritizes ethics and social responsibility
over unregulated technological advancement.

ROLE OF STAKEHOLDERS
The AI Act envisions a multi-stakeholder approach to AI
regulation, where governments, businesses, civil society, and
individuals all play critical roles. Governments are tasked with
implementing and enforcing the law, establishing national
supervisory authorities to monitor compliance, and ensuring
that AI systems meet the required safety and ethical
standards.

Companies, particularly those developing or using AI

technologies, are responsible for ensuring their systems
comply with the Act. This includes providing transparency
about their AI systems, conducting risk assessments, and
maintaining detailed documentation. Companies are also
required to put in place internal systems for monitoring and
addressing the potential risks of their AI technologies.

Civil society, including NGOs, advocacy groups, and the

general public, also plays a vital role in ensuring that AI
technologies are developed ethically. These groups often
provide oversight, raise awareness of potential risks, and
advocate for stronger protections. By engaging in public
consultations, these stakeholders help shape the regulatory
landscape and ensure that the AI Act reflects a wide range of
societal interests.

Finally, individuals who use AI systems must be aware of their

rights under the AI Act. They should be informed about how
AI systems impact their lives, whether they are being used in
hiring decisions, medical diagnoses, or personal data
processing. Individuals have the right to challenge AI
decisions that may affect them negatively, ensuring that AI
remains accountable to the public.

CASE STUDIES AND HYPOTHETICAL

SCENARIOS
In this section, we could explore real-world examples and
hypothetical scenarios to illustrate how the AI Act would be
applied in different situations. For example, a company
developing AI for medical diagnostics must demonstrate that
its system meets the strict requirements for high-risk AI,
including ensuring that it is safe, non-discriminatory, and
transparent.

A hypothetical scenario could involve an AI used by a

company to screen job applicants. Under the AI Act, this
system would need to be transparent about its decision-
making processes, undergo regular audits for bias, and
provide candidates the ability to contest decisions made by
the AI. These case studies would help showcase the real-
world applications of the AI Act’s provisions and how
companies must comply with the regulatory framework.

THE FUTURE OF ETHICAL AI

Looking ahead, the AI Act is likely to evolve as new challenges
and innovations emerge. For example, the rise of generative
AI—systems capable of producing text, images, and other
content—presents new ethical and regulatory concerns. The
AI Act will need to adapt to ensure that these technologies
are used responsibly, particularly in areas like deepfakes and
misinformation.

The EU’s investment in Horizon Europe, a funding program for

research and innovation, is another key element of the future
of AI in Europe. Horizon Europe supports the development of
ethical AI through research projects and collaborations
between governments, universities, and private companies.
As AI technology evolves, the EU is likely to adjust its
regulations to keep pace with these changes, ensuring that AI
remains aligned with European values.

Algorithm: EU Ethical AI Act Compliance

Process
START

1. INPUT: AI_System

2. DETERMINE Risk_Level of AI_System:

IF AI_System poses unacceptable risk (e.g., social scoring,
manipulative AI):
→ OUTPUT: "Prohibited under the AI Act"
→ STOP
ELSE IF AI_System is high-risk (e.g., biometrics, critical
infrastructure, employment, law enforcement):
→ GOTO High_Risk_Compliance
ELSE IF AI_System is limited-risk (e.g., chatbots, emotion
recognition):
→ GOTO Limited_Risk_Compliance
 ELSE:
→ OUTPUT: "Minimal-risk system – General obligations
apply"
→ GOTO Minimal_Risk_Compliance

-------------------------------
PROCEDURE High_Risk_Compliance:
- Perform conformity assessment
- Ensure transparency and explainability
- Maintain risk management and mitigation plan
- Set up human oversight mechanisms
- Implement data governance and bias checks
- Prepare detailed documentation
- Apply CE marking
→ OUTPUT: "High-risk AI system – Compliant"
→ RETURN

-------------------------------
PROCEDURE Limited_Risk_Compliance:
- Inform users that they are interacting with AI
- Provide opt-out options when possible
- Ensure no manipulation or emotional exploitation
 → OUTPUT: "Limited-risk AI system – Compliant"
→ RETURN

-------------------------------
PROCEDURE Minimal_Risk_Compliance:
- Ensure AI follows ethical guidelines (e.g., fairness,
transparency)
- Optional voluntary codes of conduct
→ OUTPUT: "Minimal-risk AI system – Good practice"
→ RETURN

END

Advantages of the Ethical AI Act

1. Protects Fundamental Human Rights
The Act ensures that AI systems do not violate core human
values such as dignity, privacy, freedom of expression, and
non-discrimination.
It limits the use of AI in surveillance, manipulation, and social
scoring.

2. Promotes Trust in AI
By enforcing transparency, accountability, and fairness, the
Act helps build public trust in AI technologies.
Consumers are more likely to adopt and use AI products that
are ethically regulated.

3. Defines Clear Risk-Based Regulation

The Act categorizes AI systems based on their risk
(unacceptable, high-risk, limited-risk, minimal-risk), making
the compliance process more structured and predictable.

4. Encourages Responsible Innovation

It pushes developers to integrate ethical principles from the
start of AI design, leading to more sustainable and
responsible tech development.
Ethical constraints can actually lead to more creative and
safer solutions.

5. Creates a Harmonized Legal Framework in the EU

Offers a consistent set of rules across all 27 EU member
states, preventing fragmentation and legal uncertainty.

6. Establishes Global Leadership

Through the "Brussels Effect," the EU sets a global standard
for AI governance, influencing other countries to adopt
similar regulations.
Promotes Europe as a leader in ethical AI development.

7. Introduces Accountability
Developers, deployers, and users are held accountable for
their AI systems' decisions.
There are mechanisms for complaints, redress, and penalties
for non-compliance.

Disadvantages of the Ethical AI Act

1. Regulatory Burden for Startups and SMEs
Small businesses may struggle to comply with the strict
requirements for high-risk AI systems, potentially stifling
innovation and competitiveness.

Compliance costs (audits, documentation, legal advice) can

be high.

2. Vague and Ambiguous Definitions

Some terms like “high-risk” or “human oversight” are open to
interpretation, which could cause confusion or inconsistent
application across member states.
3. Slow Legislative Process
Technology evolves faster than regulation. By the time rules
are finalized, newer AI models (e.g., generative AI, foundation
models) may fall outside the current framework.

4. Possible Overregulation
Overly strict rules may discourage companies from deploying
AI in the EU, potentially leading to "AI brain drain" or
companies shifting operations to less-regulated regions.

5. Challenges in Enforcement
Ensuring compliance across the entire EU with different
languages, institutions, and legal cultures will be complex and
resource-intensive.

National authorities may have varying capacities to enforce

the Act effectively.

6. Innovation vs. Ethics Tension

The balance between innovation and ethical governance may
tilt too far in favor of control, delaying or halting projects that
are otherwise beneficial.
Applications of the Ethical AI EU Act
1. Healthcare & Medical Diagnostics
2. Biometric Identification & Surveillance
3. Recruitment & Employment
4. Finance & Credit Scoring
5. Education & Student Assessment
6. Law Enforcement & Criminal Justice
7. Transportation & Autonomous Vehicles
8. Public Services & Government
9. Consumer Products & Smart Devices
10. Generative AI & Foundation Models (e.g., ChatGPT,
DALL·E

CONCLUSION
In conclusion, the AI Act is a pioneering piece of legislation
that establishes a comprehensive framework for the
regulation of AI in Europe. It aims to ensure that AI systems
are developed and deployed safely, ethically, and in
alignment with fundamental human rights. While the Act
faces challenges, such as concerns about over-regulation and
the need for more precise definitions, it sets an important
precedent for how AI can be governed responsibly. As AI
continues to shape the future of technology, the AI Act will
play a critical role in ensuring that AI contributes positively to
society while minimizing potential risks.