EU Parliament Bans Social Scoring and AI Tactics Exploiting User Vulnerabilities
MEPs have approved the landmark Artificial Intelligence Act, which safeguards fundamental rights, restricts harmful AI applications, and promotes innovation. The regulation, negotiated with member states and endorsed by MEPs, establishes obligations based on AI’s potential risks and impact levels. It prohibits various AI applications threatening citizens’ rights, including emotion recognition, social scoring, and manipulative AI.
As reported in europarl.europa.eu, the Act also addresses law enforcement’s use of biometric identification systems, permitting them only under strict safeguards and specific circumstances. High-risk AI systems, such as those in critical infrastructure and law enforcement, are subject to clear obligations to assess and mitigate risks, maintain transparency, and ensure human oversight. Citizens retain the right to submit complaints about AI systems affecting their rights.
Transparency requirements mandate that general-purpose AI systems meet certain standards and disclose detailed summaries of training data. Stronger GPAI models face additional requirements, including systemic risk assessment and incident reporting. The Act also mandates clear labeling of artificial or manipulated content and measures to support innovation and SMEs, such as regulatory sandboxes and real-world testing.
Co-rapporteurs emphasized the Act’s significance in reducing risks, creating opportunities, combating discrimination, and protecting workers’ and citizens’ rights. However, they underscored the need for continued work to implement the law effectively and adapt governance models to AI’s impact on societies. The regulation is pending final adoption and will enter into force gradually, with specific provisions taking effect at different intervals.
read more at europarl.europa.eu
Leave A Comment