In the race for the development of AI, China and the United States are competing for the leading role. Beijing pretends to want to go its own way. In a speech, China’s president makes demands.
Shanghai. China’s head of state and party, Xi Jinping, is demanding more control over the use of artificial intelligence and open access to the powerful technology in the AI competition with the United States. “We should strengthen risk awareness and ensure that AI is safe and controllable,” he said at the opening of the World AI Conference in Shanghai. Xi advocated a combination of laws, surveillance and early warning systems to prevent the misuse of AI.
The first speech of the president at the annual conference since 2018 shows the increased importance of KI for the ruling Communist Party. Xi campaigned for the technology to be jointly developed and made it clear that national security should not be put forward as a reason to block AI models for others. He did not mention individual countries. However, in June, the US government temporarily restricted access to certain AI models.
$(document).ready(function(){
googletag.cmd.push(function() { googletag.display(‘div-gpt-ad-1766139556442-0’); });
});
Competition with US competition
With its efforts in the technology of the future, China is ultimately challenging the AI pioneer and main rival USA. Tech companies from the United States accuse Chinese competitors of having illegally hijacked American technology in order to train their own models with it. The Chinese deny this.
A day before Xi’s speech, 29 countries signed an agreement to establish a World Organization for Cooperation on Artificial Intelligence (WAICO) based in Shanghai. According to the state-run Chinese news agency Xinhua, the signatories include well-meaning countries such as Russia, Indonesia and Kazakhstan. WAICO should therefore promote cooperation and global governance in the field of AI.
China would thus have a platform to negotiate standards in the field of AI. In any case, Beijing has long demanded that the development of artificial intelligence should not be in the hands of a few countries. “The future of AI should not be determined by monopolies or geopolitical rivalries, but by the extent to which its innovations are shared and how widely its benefits are felt,” commented Xinhua, which is considered the mouthpiece of the Communist Party.
China startup presents new AI model
Just in time for the start of the conference, the Chinese startup Moonshot AI published a new AI model with “Kimi K3”, which according to the company should have 2.8 trillion parameters. The number of parameters indicates the complexity of the model. Simply put: the more parameters it has, the higher the performance.
The Beijing company thus wants to catch up with the US rivals OpenAI and Anthropic, which have so far solved particularly complex tasks better with their most powerful models. The two companies do not make their parameter numbers public. However, market experts expect that the lead of the US developers over the ever stronger and cheaper AI models from China could melt.
Source: Tageblatt
