The European Union (EU) is now the first union to implement a comprehensive legal framework on the use of AI. These regulations plan to target the societal risks associated with AI systems, particularly those believed to pose significant threats to fundamental rights, safety, and ethical principles. One of the key objectives is to address the lack of explanations in AI decision-making processes. By requiring AI systems to provide the why, how, and what of their decisions, the Act aims to limit the risk of unfair disadvantage or discrimination, ultimately promoting fairness and accountability in AI-driven processes.
For example, consider the use of AI in vetting resumes for a potential job. In a recent study conducted by Bloomberg, it was determined that OpenAI’s generative technology displayed racial preferences in the employment process. AI decisions lack transparency, ultimately leading to unjust outcomes such as unfair disadvantage in hiring processes or public benefit schemes. It discriminated against ethnic sounding names while pushing forward American names. Implementing these new rules was vital to protect privacy and minimizes discriminatory based use.
This also created a productive environment for businesses to develop and deploy AI solutions—including measures such as reducing administrative and financial burdens for businesses while ensuring compliance with regulatory requirements. With a clear framework of what they can and can not do, it is easier for them to safely use AI within their workspace.
Some businesses look towards AI in public surveillance. The act proposes restrictions on real-time biometric surveillance technologies, like facial recognition, in everyday public spaces. It is essential to ensure that our personal information, including our faces, isn’t being collected and analyzed without our consent. By restricting the use of facial recognition in public spaces, the AI Act helps prevent potential abuses of surveillance technology and protects our right to privacy and ensures that these tools are only utilized in exceptional scenarios, like apprehending dangerous criminals.
Furthermore, the Act provides a guideline for “low risk” systems. For example, virtual assistants in the home is a technology that has become widespread use of AI over the course of a few years. These virtual assistants, like Siri or Amazon Alexa, are designed to perform simple tasks and provide users with information or assistance in various daily activities, therefore categorized as “low risk.” These types of systems pose minimal harm to society as a whole. Their functionalities are limited to tasks such as setting reminders, answering questions, playing music, controlling smart home devices, etc. These applications do not involve high-stakes decision-making or sensitive data processing.
The EU’s AI Act is a huge step forward in the right direction. It lays the foundation for a future where AI serves as a tool for progress, not a threat to our freedoms. The ethical development and use of AI is no longer a choice, but a global responsibility. With this legislation, Europe sets the stage for a global conversation, inviting other nations to collaborate and build a future where AI is used responsibly.
Written by Destiny Anna
