STOCKHOLM, April 27 (Reuters) – Companies deploying generative AI tools like ChatGPT must disclose any copyrighted content to develop their systems, which is already under an EU agreement governing the world’s first general set of rules for the technology.
Two years ago, the European Commission began drafting AI legislation to regulate new artificial intelligence technology that has grown in investment and popularity following the release of OpenAI’s AI-powered chatbot ChatGPT.
Members of the European Union Parliament have agreed to move the draft to the next stage, which is to log three; During this period, EU legislators and member states released the final details of the law.
According to the recommendations, AI tools are classified according to the level of perceived risk: from low to limited, high and unacceptable. Concerns may include biometric tracking, misinformation, or discriminatory language.
Even if high-risk devices are not banned, those who use them must be very transparent in their work.
Companies deploying generative AI tools like ChatGPT or image generator Midjourney must disclose any copyrighted material used to develop their systems.
A source familiar with the discussions said that this provision was added late in the last two weeks. Some committee members initially proposed that copyrighted material be banned from training generative AI models in general, the source said, but that left a requirement for transparency.
According to Svenja Hahn, a deputy of the European Parliament, “conservative aspirations for more surveillance and nightmares of over-regulation, the Parliament found a strong agreement to regulate AI in a fair way, protect the rights of citizens, as well as encourage innovation and grow the economy”.
Macquarie analyst Fred Havemeyer said the EU’s proposal is a “method” rather than a “ban first and ask questions later” approach.
“The EU was on the borderline in regulating AI technology,” he told Reuters.
Market competition
Microsoft-backed ( MSFT.O ) OpenAI sparked fear and anxiety around the world when it unveiled ChatGipt last year. The chatbot became the fastest-growing consumer application in history, reaching 100 million monthly active users in a matter of weeks.
The race among tech companies to bring AI products to market has worried some observers, with Twitter owner Elon Musk backing a proposal to halt development of these systems for six months.
Shortly after signing the letter, the Financial Times reported that Musk was planning to launch his own startup to rival OpenAI.
The report was written by Supantha Mukherjee in Stockholm and Fo Yun Chi in Brussels
Our standards: The Thomson Reuters Trust Principles.
We offer you some site tools and assistance to get the best result in daily life by taking advantage of simple experiences