European Union’s AI Act – The Dawn of a New Regulatory Regime
The European Union’s Artificial Intelligence Act (AI Act) stands as a pivotal milestone in the European Union’s pursuit of effective artificial intelligence (AI) governance. This groundbreaking proposal strives to establish a comprehensive regulatory framework that harmonizes the development, market deployment, and responsible usage of AI systems throughout the EU. It is the first comprehensive regulatory framework for AI systems in the world. The AI Act is expected to significantly affect the governance of AI systems in the EU. It is also expected to have a broader impact as other jurisdictions worldwide consider the AI Act a model for their own AI regulations.
AI Regulation In India – A Good Start Is Half The Battle
The Indian government has not yet announced plans to introduce a specific law to regulate AI. However, the government has taken a few steps to promote the responsible development and use of AI. For example, in 2020, the government published a national strategy for AI that sets out principles for the responsible development and use of AI. The government has also established several working groups to study AI’s potential risks and benefits and develop recommendations for mitigating the risks.
The Indian government might introduce a specific law to regulate AI. However, the government also might continue to take a more informal approach to regulating AI by issuing guidelines and best practices.
Overview of the AI Act
The primary objective of the AI Act is to safeguard the development and utilization of AI systems in a manner that prioritizes safety, reliability, and responsibility while simultaneously upholding fundamental rights and freedoms.
The AI Act classifies AI systems into three distinct risk categories. AI systems categorized as presenting an unacceptable level of risk, such as those employed for social scoring, mass surveillance, or biometric identification without consent, are prohibited. High-risk AI systems, which encompass those utilized in automated decision-making, critical infrastructure, or law enforcement, are subject to specific requirements, including mandatory conformity assessment and market surveillance. Conversely, low or negligible risk AI systems are not subjected to any particular requirements.
In addition to the risk-based classification, the AI Act sets forth general requirements that are applicable to all AI systems, regardless of their risk category. These requirements encompass transparency, accountability, robustness, and privacy. AI systems must be designed to ensure transparency and comprehensibility, providing users with information concerning their functionality, training, and decision
making processes. Furthermore, provisions must be in place to ensure accountability, allowing for human intervention when deemed necessary. It is imperative that AI systems are designed in a manner that guarantees robustness, resilience against manipulation, bias, and errors. Additionally, AI systems must uphold privacy standards and comply with the General Data Protection Regulation (GDPR) when handling personal data.
The AI Act is still under negotiation, but it is expected to be adopted by the European Parliament and the Council of the European Union in 2023. Once adopted, the AI Act will directly apply to all EU Member States.
Key Takeaways
- The AI Act demonstrates the EU’s commitment to responsible regulation of AI, setting a pioneering example for the global community. The AI Act has garnered acclaim from experts who commend its ambitious and comprehensive nature.
- Clear regulations and guidelines are necessary to ensure the safe, responsible, and ethical development and utilization of AI. The AI Act serves as a foundation in this regard, facilitating the promotion of responsible AI practices. Nevertheless, some experts have expressed that the AI Act could be more extensive. Specifically, they propose the inclusion of more explicit mandates pertaining to AI systems implemented in sensitive domains like law enforcement and healthcare.
- The AI Act could serve as a compelling model that can be emulated by other countries seeking to advance their own frameworks for AI regulation and governance.