As we step deeper into the age of Artificial Intelligence (AI), it’s evident that AI is no longer just a futuristic concept. From our daily routines to sectors like education, entertainment, and healthcare, AI technology is becoming increasingly integrated in our lives, transforming the way we live, learn, work, and interact. It’s already influencing decisions of industries, shaping products and services, and impacting the way we make choices, both big and small.
On February 2, 2024, the Council of the European Union (EU) took a historic step by unanimously approving the EU AI Act, a landmark piece of legislation that will guide the responsible development and use of AI across Europe. The Act officially entered into force on August 1, 2024, but the application of its various parts is phased out. On February 2, 2025, just a few months away from now, the first two chapters (Article 4 and 5) will come into effect!
In this article, we’ll explore the EU AI Act, review the first two chapters, and, importantly, share how your business can prepare for this upcoming legislation.
What is the EU AI Act?
AI brings vast opportunities for innovation, driving advancements in both human and industrial development. But it also raises complex issues and challenges around privacy, transparency, and accountability. When AI is used in wrong hands, it can potentially become a “cyberweapon”, as described by the World Economic Forum.
In response to these challenges, the EU has decided to establish clear rules around AI applications, offering a robust and future-proof regulatory framework. From the development of AI technology to its applications across EU member states. The EU AI Act takes a risk-based approach, meaning that it classifies AI applications into four risk categories. Each risk category comes with distinct rules and obligations:
- Unacceptable Risk: AI applications deemed as harmful to society, such as social scoring systems, are strictly prohibited.
- High Risk: This includes AI systems used in sectors like healthcare, transportation, and law enforcement, which could significantly impact human rights or safety. These systems are allowed but subject to strict regulations and compliance requirements.
- Limited Risk: Applications in this category require some level of transparency but are subject to less stringent regulations.
- Minimal Risk: The least restricted category, which includes most AI applications, such as chatbots and recommendation algorithms, is not regulated under the EU AI Act.
The EU AI Act first two chapters: Article 4 & 5
As mentioned above, the first two chapters will come into effect in 3 months. Basically, Article 4 and Article 5 outline critical regulations that directly address ethical concerns in AI usage and impose specific prohibitions on certain AI applications.
Article 4: AI Literacy
This article emphasizes the need for AI literacy and education among developers, users, and operators. Companies are required to ensure that employees and users operating or interacting with AI systems have sufficient knowledge and training on how to use these systems responsibly and safely. This mandate seeks to ensure that AI is understood and handled effectively, reducing misuse and promoting transparency in the use of AI tools.
Article 5: Prohibited AI Practices
Article 5 identifies several high-risk AI practices that are strictly prohibited to protect fundamental rights and prevent harm. These prohibited applications include:
- Subliminal manipulation: AI systems that use subliminal techniques to influence individuals’ behaviors without their awareness, potentially causing harm.
- Exploitation of vulnerabilities: Systems that exploit individuals’ vulnerabilities based on characteristics such as age or disability to manipulate behaviors.
- Social scoring: Public authorities are restricted from using AI to create social scoring systems that evaluate individuals based on their behavior or personal traits, as this practice could lead to unjust discrimination.
- Predictive policing: AI systems that assess individuals’ risk of committing crimes based solely on personal characteristics are banned, although they may support
How can your business prepare?
With the EU AI Act’s first phases imminent, businesses — especially small and medium-sized enterprises— may feel pressure to comply. However, this regulation also presents an opportunity to stand out as a leader in ethical AI.
To prepare:
- Build AI Literacy: Ensure your team understands and handles AI systems responsibly (Article 4).
- Audit practices: Review your AI applications to confirm they don’t fall into prohibited categories (Article 5).
- Develop compliance documentation: Prepare records and monitoring practices that demonstrate your commitment to AI regulations.
Unlock AI excellence with Swiss Cyber Institute
At the Swiss Cyber Institute, we’ve launched a comprehensive series of AI courses designed to help you develop expertise and succeed in this dynamic field.
Our program addresses the needs of previous students who sought a structured pathway to deepen their understanding of AI while keeping pace with the current landscape. Through our courses, you’ll gain hands-on skills and insights into the latest advancements, empowering you to harness the full potential of AI in your industry.
AI Literacy
Our AI Literacy course provides a comprehensive introduction to foundational concepts, essential technologies, practical AI tools, and the ethical considerations surrounding AI. It requires no prior knowledge or experience, making it accessible to all learners.
To be complied with the EU AI Act Article 4, this training is a perfect chance for employees to be equipped with sufficient knowledge and hands-on training in the use of AI technology.
AI Prompt Engineering
The objective of our AI Prompt Engineering course is to master AI prompt engineering and excel in the AI field. This training focuses on more specific practical skills for multiple industries, including healthcare, finance, and technology that utilizes AI and NLP. This training emphasizes practical applications, equipping participants with the expertise to stand out in creating effective AI prompts and aligning with the EU AI Act’s principles of transparency and accountability.
If you already have foundational knowledge of AI, and ready to advance your skills further, this training provides an excellent opportunity for growth in the field as well as for the compliance of the EU AI Act for your business.
Start preparing now with our AI training series to ensure your team is equipped to navigate AI responsibly. Ready to stay compliant and thrive in an AI-driven future? Contact the Swiss Cyber Institute today.
