Dealing with the complexities associated with complying with extensive regulations, such as the EU’s Artificial Intelligence Act, creates significant challenges for organizations of all sizes.
At NeuroSYS, we recognize these difficulties, so we have created an AI Compliance service that ensures compliance with current regulations. With the emergence of a new rule under the EU Artificial Intelligence Act, we decided to provide you with the most essential information from the Act.
Stay current with the EU AI Act to ensure your systems comply with evolving regulatory requirements.
Understanding the EU Artificial Intelligence Act
The EU Artificial Intelligence Act is a proposed regulatory framework initiated by the European Commission on April 21, 2021. It achieved political consensus across all three EU institutions by December 8, 2023.
This regulation establishes a detailed framework of rules for both providers and users of AI systems. These rules specify each party’s transparency and reporting responsibilities when introducing AI systems into the EU market.
The EU AI Act is poised to apply to European entities and any AI system that impacts EU citizens. This includes companies introducing AI systems to the EU market or those whose AI system outputs are utilized within the EU, regardless of these systems’ development or deployment location.
While the primary provisions of the EU Artificial Intelligence Law were agreed upon on December 8, certain technical specifics are yet to be finalized. Specialized teams are in the process of completing these details.
The Parliament’s Committee on the Internal Market and Civil Liberties is scheduled to vote on the final agreement, which is expected on January 25, 2024. Following the parliamentary vote, the finalized text of the EU’s AI Law is anticipated to be published in the EU’s Official
The publication of the text will mark the commencement of the enforcement deadlines.
Comprehensive Fines Structure in the EU Artificial Intelligence Act
The EU AI Act introduces a stringent fines structure to ensure compliance and responsible use of AI technologies. The proposed fines vary depending on the severity and nature of the violation:
Severe AI Violations:
For the most severe breaches, particularly those involving prohibited AI practices, organizations could face fines of up to 7% of their global annual turnover or a penalty of €35 million, whichever is higher. This category includes violations that pose significant risks to individuals’ safety, rights, and freedoms.
General AI Compliance Violations:
For other types of non-compliance, which may include failure to adhere to mandated AI system standards or operational guidelines, fines could reach up to 3% of the global annual turnover, or €15 million, whichever is greater. This level addresses a broad range of violations, reinforcing the importance of comprehensive AI governance.
Misrepresentation or Incorrect Information:
In cases where entities provide false or misleading information about their AI systems, they could be fined up to 1.5% of their global annual turnover, or €7.5 million. This ensures accountability and transparency in AI system reporting and documentation.
Special Considerations for SMEs and Startups:
Recognizing the potential impact on smaller businesses, the Act includes provisions for capped fines for Small and Medium-sized Enterprises (SMEs) and startups. This is intended to balance the need for regulatory compliance with the economic realities of smaller AI players.
The structure of these fines underscores the EU’s commitment to ensuring safe and ethical AI development and deployment while also considering the diverse nature of businesses operating in this space. This framework aims to penalize non-compliance and encourage proactive adherence to high standards of AI ethics and safety.
Responsibilities of Entrepreneurs Under the EU Artificial Intelligence Act
Entrepreneurs developing or utilizing AI systems are responsible for aligning with the EU AI Act. This period should be used for preparation and ensuring compliance.
The extent of compliance duties hinges on the risk level posed by the AI system to human safety and fundamental rights across the AI value chain.
The EU AI Act introduces a layered compliance framework, primarily focusing on AI systems categorized as “high risk” and general-purpose AI systems, including core and generative models, which carry significant impact and systemic risk.
Your responsibilities, contingent on your system’s risk level, might include:
- Risk Assessment:
Evaluating the risk associated with your AI system to classify its level of risk.
- Conformity Assessment:
Providing an assessment of conformity, either through self-assessment using EU-endorsed technical standards or via a third-party evaluation by an EU-accredited body.
- Documentation and Record Keeping:
Keeping up-to-date technical documentation and records for your AI systems.
- Transparency and Disclosure:
For Prohibited AI: No transparency obligations are required as these systems should be withdrawn from the market.
High-Risk AI: Registering high-risk AI systems in the EU database before market introduction.
Mitigated Risk AI: In cases of permitted emotion recognition or biometric categorization systems, inform and obtain user consent. AI-generated “deepfake” content must be revealed and clearly labeled.
Minimal Risk AI: No specific transparency obligations are mandated.
This framework aims to ensure that entrepreneurs comply with regulatory standards and maintain ethical practices in AI development and deployment.
Find current EU AI Act regulation HERE
Leveraging NeuroSYS for EU Artificial Intelligence Act Compliance
NeuroSYS emerges as a strategic ally in your journey to comply with the EU Artificial Intelligence Act. Our expertise in AI technology positions them perfectly to assist your organization in understanding and implementing the required compliance measures. We offer tailored solutions that help identify the risk categories of your AI systems, aligning with the Act’s nuanced compliance framework.
Take advantage of a free one-hour consultation to discuss your needs related to implementing AI in your company’s organizational structure.
