“Hiroshima Process”: G7 countries want to regulate generative AI more closely
The representatives of the G7 countries have agreed that generative artificial intelligence (AI) must be regulated within the framework of new standards. Attendees expressed concern about the technology’s rapid development and disruptive potential. As part of the so-called “Hiroshima Process”, proposals for regulating ChatGPT and Co. should be available by the end of this year.
AI must be humane and trustworthy
In Hiroshima, the major industrialized countries fear powerful tools for disinformation and political disruption as AI-driven models for text, images, and video develop rapidly. For this reason, the governments of the G7 countries should discuss possible framework conditions for the limitation and regulation of artificial intelligence at cabinet level and present the results by the end of the year.
When developing AI, it must be ensured that it is humane and trustworthy, demanded Japanese Prime Minister Fumio Nishida at the G7 meeting in Hiroshima (Japan). Nishida also pledged financial support for efforts to work together to ensure secure cross-border data flow. According to Bloomberg, however, Japan’s government prefers less strict guidelines for AI than those required by the EU.
Detailed laws vs. technical changes
It’s important to involve as many countries as possible in the AI regulation discussion, including lower-income countries, said Kyoko Yoshinaga, senior fellow at the Institute for Technology Law & Policy at Georgetown University Law Center. Hiroki Habuka, a senior associate at the non-profit Wadhwani Center for AI and Advanced Technologies in India (Mumbai), explains the importance of governments ultimately cracking down on major issues with tough laws. However, he warned that laws that were too detailed would not be able to keep up with the technical changes.
The demands for particularly strict regulation for ChatGPT and Co. had already been called for by MEPs in April and even OpenAI and ChatGPT boss Sam Altman pleaded for stricter regulation at a hearing in the US Senate. He compared the possibility of manipulated text with that of images and described his own company’s language model as “like Photoshop on steroids”.
G7 countries increase pressure on China
In addition to the AI discussion, the industrialized countries agreed on a tougher approach towards China. The risks in China business and for the supply chains must be reduced. China must abide by the rules, create fair competitive conditions in its market and reduce risks in dealing with China and diversify supply chains.
The G7 countries want to turn against “economic coercive measures”. One sees with concern that China often uses its economic weight in political disputes and puts pressure on other countries. “We want to reduce the risk of not disconnecting,” said US President Joe Biden’s National Security Advisor Jake Sullivan. In the run-up to the summit, there were definitely differences about the right way to go about dealing with China. Nevertheless, one recognizes that each country maintains its own relationship with the People’s Republic.
Investment review to protect sensitive technologies
In order to protect sensitive technologies, the G7 states want to subject investments abroad (which are primarily aimed at China) to a more detailed preliminary examination – with a view to economic security. After that, each country will have to decide for itself how to approach the issue, according to US National Security Advisor Sullivan. Accordingly, a country should not pour large amounts of capital into industrial sectors of rival states that are important for the national security of its own country – for example in the defense or technology sector.
Despite all the differences with China, the G7 states underline their interest in stable and constructive relations – China must be involved in global challenges and also allow itself to be involved. They want to work with China on issues such as climate change and financial stability.
(bme)
To home page
