The rapid development of artificial intelligence (AI) technology has stimulated emergency dialogue on powerful regulations. Professor Paul Dumouchel of Quebec University in Montreal believes that the general misunderstanding of the nature of AI has a significant impact on how to regulate. The article “AI and Regulations” was published in AI journals.
Professor Dumouchel emphasized that AI should not be regarded as a single, overall entity. Instead, it includes a variety of cognitive technologies with different characteristics and applications. This misunderstanding usually leads to low adjustment methods. He said: “Seeing artificial intelligence as something that exists, not as a set of cognitive technologies … the way to lead to low supervision efficiency.” This view emphasizes the multi -faceted nature of AI. The importance of mechanism.
With the emergence of advanced AI models such as ChatGPT, the urgency of AI regulations has increased. Ask questions: Should I supervise AI? If so, what and who? Professor Dumouchel explored these issues and pointed out that the diversity of AI technology complicated the regulatory landscape. He explained that AI covers various devices, algorithms and applications, and each device needs customized supervision methods.
Professor Dumouchel traced the historical root of AI regulation and pointed out that the early AI research was based on the following assumptions: human thought can be accurately copied by machines. This has caused people to believe that artificial intelligence is fundamentally similar, which affects regulatory thinking. However, Professor Du Muchel believes that this view is defective. He wrote: “There is no difference between claiming the performance of human thought and artificial cognitive system, so it depends on the possibility of creating machines that can copy various knowledge operation results.”
He criticized two common methods of AI adjustment. The first is the proposal to suspend the AI research. He describes it as a contradiction and ultimately invalid measure. Professor Dumouchel asserted: “But suspension is not a form of supervision. It is more like a double binding. This is a contradiction that is destined to fail.” The second method involves embedding moral considerations into the AI system, He believes that this ignores the fundamental difference between humans and artificial agents.
Instead, Professor Dumouchel advocates a regulatory framework that focuses on the unique features of the AI system. He emphasized the importance of understanding the specific cognitive field and limitations of AI proxy. Unlike humans, the AI system is limited to data and algorithms that define its functions. This difference is essential to formulate regulations to solve the actual function and risk of AI.
He also emphasized the importance of transparency and accountability in the AI system. Because the AI agent runs intangiblely, they must ensure that their actions can be traced and understandable. Professor Dumouchel pointed out: “We have never encountered an agent itself because it is ultimately a mathematical function. We only face some consequences of its functions.” Therefore, the regulations should request a clear report and accountability mechanism for the AI system.
Professor Dumouchel’s conclusion is that the main challenges of AI regulations are politically, not moral or metaphysical. The deployment of artificial intelligence technology has changed the dynamics of power among different social participants, and supervision methods need to consider these transformations. He wrote: “The artificial cognitive system and related information and communication technologies have always been the core in the unprecedented wealth and power concentration we have witnessed.”
All in all, Professor Du Muchel’s research emphasizes the subtle and the necessity of AI supervision. By regarding AI as a variety of cognitive technologies, decision makers can develop a more effective supervision mechanism to solve the specific risks and capabilities of the AI system. Transparency, accountability system, and understanding of AI limitations are an important part of a powerful regulatory framework.
Journal reference
Dumouchel, Paul. “AI and regulations.” AI, volume. 4, no. 4,2023, PP. 1023-1035. Doi: https: //doi.org/10.3390/ai4040052
About the author
Paul Dumouchel It was Canadians, until the recent core morality and the graduate school of border science, his Japanese Kyoto Kyuster University, he believes in political philosophy and scientific philosophy. He is mood (Seuil, 1999) The contradiction between scarcity and other papers (2014) and Barren sacrifice (2015) all at Michigan State University Press. He edited Opposition to unfair: Amariianson’s new economics (Cambridge University Press, 2009) and Social bond as freedom (Berghahn Books, 2015). His book with Luisa Damiano recently is Vivre Avec les robot (Seuil, 2016) English translation Live with robots (Harvard University Press) He was distributed in Italian and South Korea in 2017 and 2019.
Paul Dumouchel E., 2003-2021 Professor Ilétait De Poliesophie Politique et De Philosophie Des SCIENCES -Core Ethical Graduate School and Border Science Graduate School Del’ustryérittSumeikan -K IOTO AU JAPON. IL Est Avec Jean-Pierre Dupuy Co-AUTEUR de L’enfer des select (Seuil, 1979). IL Est Aussi Audeur de Émotions (Seuil, 1999) AINSI QUE de LE sacrificed invalid (Flammarion, 2011) Demêmeque The contradiction between scarcity and other papers (2014). Avec Reiko Gotoh Il Est co -directed de de de de Opposition to unfair: Amariianson’s new economics (Cambridge University Press, 2009) AINSI QUE De Social bond as freedom (Berghahn Books, 2015). Son, Livre Le PlusrécentaveCluisa Damiano, Vivre Avec les robot (Seuil, 2016) A An Anglais, EN Italien et Encoréen. Il EsterEMENT Professor AUDépartementde L’Iversophie Del’Insionitéduquébecàmontréal, Canada.
