Meta refuses to sign EU’s AI code of practice

Meta refused to sign the European Union Experience code For the AI Law, a few weeks before the rules of the block for the providers of the general models of AI models.

“Europe is on the wrong way in the AI,” said Meta’s Chief Global Affairs Officer Joel Kaplan wrote post In LinkedIn. “We will not sign a general purpose of European Commission (GPAI) models and meta. This code presents models, as well as a large number of legal uncertainty within the AI Act.”

The EU Experience Code – a voluntary framework that published this month aims to help companies implement processes and systems to comply with the legislation of blocks to regulate the AI. Among other things, the code requires companies to provide documents on AI tools and services and regular updates; Those who prohibit AI training in pirate content; and in line with the requirements of the contents do not use their work in the data sets.

Calling to implement the EU legislation to implement the legislation of the law, “European companies who think that border AI models in Europe will develop and develop business on them,” he said.

One Risk-based regulation for applications of artificial intelligenceAI ACT, as a cognitive behavioral manipulation or social goals, using some “unacceptable risk”, using some “unacceptable risk”, uses cognitive behavior manipulations or social goals. The rules are also a biometric and domains such as “high-risk” as biometric and face recognition. The act also requires the development of developers to register the AI systems and fulfill the obligations of risk and quality management.

Technological companies from all over the world, which are in the forefront of the AI race such as Alphabet, Meta, Microsoft and MistroSoft and Mistral AI we battle The rules even call on the European Commission to delay. But there is a commission fortifiedsaid it would not change the schedule.

EU was published in the EU leadership For providers of AI models before the rules to enter into force on August 2. These rules will affect the providers of “Systematic Risk General AI models” such as Openai, Anthropic, Google and Meta. Companies with such models in the market before August 2 must be in accordance with the legislation until August 2, 2027.