Summary:
L’Union européenne envisage de désigner ChatGPT comme un service réglementé en vertu de la Loi sur les services numériques (LSN). Ce mouvement vise à répondre aux risques systémiques posés par les services en ligne alimentés par l’IA dépassant les seuils d’utilisateurs fixés par la LSN. OpenAI a révélé que la fonction de recherche de ChatGPT a en moyenne plus de 120 millions d’utilisateurs mensuels dans l’Union européenne au cours des six derniers mois, dépassant largement le seuil de 45 millions d’utilisateurs de la LSN; si cela est confirmé, ChatGPT deviendrait le premier service d’IA autonome réglementé en vertu de la LSN, établissant un précédent pour des services similaires et s’alignant sur les prochaines règles de la Loi sur l’IA de l’UE.
Original Link:
Generated Article:
The European Union’s Digital Services Act (DSA) may soon classify ChatGPT as a regulated service due to its significant user base within the EU. According to OpenAI, the ChatGPT search feature averaged over 120 million monthly users across the EU in the past six months, surpassing the 45 million threshold established by the DSA for services that may pose “systemic risks.” This marks a pivotal moment, potentially making ChatGPT the first standalone AI-driven service subject to regulatory oversight under this framework.
The Digital Services Act, which came into effect in November 2022, aims to create a safer and more accountable online environment by outlining obligations for digital platforms and services operating in the EU. It identifies thresholds for platforms that should be subject to heightened scrutiny for consumer protection and public safety reasons. Services crossing the threshold of 45 million average monthly EU users—a metric reflecting significant societal dependence—are classified as “very large online platforms” or “very large online search engines.” Once designated, these platforms are required to take robust measures to address illegal content, ensure transparency, and assess potential systemic risks, including threats to fundamental rights like privacy and freedom of expression.
ChatGPT’s possible regulation under the DSA raises ethical considerations about balancing innovation against the need for accountability. AI systems like ChatGPT can potentially propagate harmful misinformation, discriminatory biases, or violate privacy. Regulation under the DSA would impose duties on OpenAI to evaluate and mitigate these risks in accordance with EU standards, thus establishing an important precedent for future AI advancements. However, some may argue that stringent regulations could stifle creativity and hamper technological growth.
The implications for the tech industry are significant. First, OpenAI’s designation under the DSA would signal to developers that AI systems drawing substantial user engagement are no longer operating in a regulatory vacuum. This could lead to increased compliance costs as companies navigate the process of risk assessments, audits, and the payment of supervisory fees, such as the one the European Commission is calculating in coordination with Ireland’s Media Commission. On the other hand, it could also bolster consumer trust in AI systems as these technologies become more transparent and accountable under regulatory frameworks.
Similar efforts to govern AI products are already underway through the EU AI Act, which focuses specifically on high-risk AI systems and their potential societal impacts. Should ChatGPT’s designation be confirmed, it may provide clarity on how these two regulatory schemes—the DSA and the EU AI Act—will intersect in governing AI-driven services. By acting as the first regulated standalone AI service, ChatGPT could shape compliance standards for other AI platforms, influencing not just EU policy but also global approaches to AI regulation.
For instance, consider potential changes in application use: if OpenAI is required to make significant algorithmic adjustments to meet DSA guidelines, users may experience changes to how ChatGPT processes and outputs information. These modifications could set benchmarks for competitors, creating a ripple effect across the industry.
As society and policymakers continue grappling with the ethical and practical implications of AI, the regulation of services like ChatGPT underscores a crucial transition—AI technologies are not just tools but also entities capable of widespread influence, demanding oversight to ensure alignment with public interest and societal norms.