Artificial intelligence (AI) has been a part of our lives for many years and continues to make headlines with ever-smarter multitools. One of the most well-known is ChatGPT from OpenAI. With more than 100 million monthly users just two months after its launch, the chatbot officially holds the record for the fastest-growing web platform in history the use of ChatGPT.
But despite all the excitement, ChatGPT, like almost all electronic tools, also raises questions about its impact on data protection and cybersecurity . Depending on the questions or tasks posed to ChatGPT, users disclose varying amounts of information, some of which is sensitive—for example, about interests in political, religious, ideological, or scientific issues, as well as their family or sexual situation.
Can the principles of the General Data Protection Regulation be compli with? the use of ChatGPT
According to the General Data Protection Regulation (GDPR), a provider of services that process personal data must, among other things, be able to explain which data is process, how, and for what purpose. When using ChatGPT, however, usa business fax list it is unclear for what purposes the entered data is process and from which data pool the artificial intelligence behind the service draws its knowledge. Here, the principle of transparency is at odds with a self-learning system that produces only partially comprehensible results.
The principle of data minimization (storage limitation) also conflicts the largest possible amounts of data for effective training. Under the GDPR, the controller is also obligatto ensure the accuracy of the personal data it processes. However, the algorithm’s knowledge may become outdated if circumstances change after its training, even if the original data was correct.
Reactions of data protection authorities in Europe
The popular AI tool ChatGPT was temporarily blocked by the Italian data protection authority at the end of March 2023 because it was deemed incompatible with data protection regulations. Despite the subsequent improvements in data protection made by the operator through new settings, concerns remain. In Europe, other countries have announced or initiated investigations relating to the compatibility of ChatGPT and the GDPR, including Spain, how to buy a com.tr domain name? what documents are required? France, and Ireland. The German supervisory authorities in the Conference of Independent Data Protection Supervisory. In order to be able to provide an assessment, they first asked the San Francisco-based company OpenAI to answer a questionnaire about data processing at ChatGPT. According to the German Federal Data Protection Commissioner, a blocking in this country is also conceivable.
Regulations are urgently need. But what should they look like?
(AI Act) since April 2021 , and it could be pass before the end of this year. The law is intend to regulate the use of artificial intelligence in Europe and aims in particular to improve data quality, transparency, human oversight, united states business directory and accountability in the field of AI. The draft law divides AI applications into risk classes. Low-risk programs will be barely regulated, while overly risky systems are to be banned altogether. These include, for example, social credit systems that states could use to analyze the everyday behavior of their citizens. The third class consists of high-risk applications. Their providers must comply with requirements to minimize these risks. For example, an AI that suggests treatment methods in a hospital would be such a high-risk case.
Numerous voices advocate for strict regulations for manufacturers of general-purpose AI systems in the AI Act. The current draft of the EU AI law would constitute overregulation.
What the future holds
Legislative regulations such as the AI Act are intend to strike a balance between technological advances and individuals’ rights to privacy and information security. Whether this will succeed in practice remains to be seen.