Beijing has indicated that it acknowledges Washington’s decision to use the term “super-intelligence”, or “SI”, in its official communications instead of the term “artificial intelligence”. China says it respects this choice of vocabulary, without confirming that Xi Jinping approved it during his meeting with Donald Trump. The two countries also announced the creation of a communication channel dedicated to incidents related to AI.
The Chinese statement came on Saturday, September 26, after American President Donald Trump announced a change in terminology for U.S. government communications. According to this announcement, the term “artificial intelligence” was to be replaced by “SUPER INTELLIGENCE”, abbreviated as “SI”.
The Chinese Ministry of Foreign Affairs stated that Beijing takes into account the American position and respects the formulation chosen by Washington. This measured response neither constitutes an explicit endorsement of the new term nor a challenge to the decision. It mainly indicates that China recognizes the terminological choice of its interlocutor while leaving the discussion open on how to name and frame these technologies.
The spokesperson for the ministry also emphasized that technologies related to AI continue to evolve. According to him, the various parties can take these advancements into account to strengthen their exchanges, deepen their discussions, and seek areas of agreement. This wording places the issue of vocabulary within a broader framework of technological dialogue between the two powers.
A designation highlighted by Donald Trump
The American account of the meeting with Xi Jinping
Donald Trump mentioned this change after his meetings with Chinese President Xi Jinping, organized in Washington during the week. On his social media platform, he claimed that his counterpart seemed to appreciate the replacement of the term “artificial intelligence” with a name he deemed more accurate and important.
Beijing has not explicitly confirmed this interpretation. The official Chinese response addresses the choice of American terminology but does not validate the account that Xi Jinping personally approved the term “super-intelligence.” This distinction leaves a gap between Donald Trump’s comment on his counterpart’s reaction and the position publicly stated by Chinese authorities.
A change in official communications
The American announcement concerns how the U.S. government intends to designate this technology in its communications. It is primarily a decision related to political and institutional vocabulary. The replacement of one term with another does not, by itself, change the technical capabilities of existing systems, their level of development, or their degree of autonomy.
The choice of the term “super-intelligence” does, however, give a different tone to the debate. Whereas “artificial intelligence” serves as a general term for a wide range of systems and methods, “super-intelligence” evokes a form of ability that is far superior to human capabilities. The designation can thus influence the way the public understands the issues, even if it does not constitute a precise technical definition of all tools grouped under the name of AI.
The distinction between artificial intelligence and superintelligence
A more restricted technical term
In research, superintelligence generally does not refer to the entirety of artificial intelligence technologies. The term rather refers to the hypothesis of a system whose cognitive capabilities would exceed those of a large number of human beings, or even those of vast human organizations in various fields.
This notion must be distinguished from general artificial intelligence, often referred to by the acronym AGI. This is generally described as an artificial intelligence capable of performing a wide range of tasks at a level equal to or greater than that of a human. Definitions vary among researchers and organizations, but AGI and superintelligence are not synonymous: the latter evokes a level of capabilities far exceeding human level.
Currently available systems are not, by virtue of their designation or progress, considered to have reached the level typically associated with superintelligence. The decision from Washington thus translates into a change in official language, rather than a recognition that current technologies have crossed this hypothetical threshold.
Words that shape public debate
The terms used by political leaders can influence expectations and concerns around innovation. Talking about “super-intelligence” emphasizes potentially extraordinary capabilities and scenarios of surpassing human skills. In contrast, the term “artificial intelligence” encompasses both specialized tools and generative systems or decision aid devices.
This difference in scope is important when officials discuss risks. Donald Trump has repeatedly downplayed the notion that artificial intelligence could pose a threat to humanity. Xi Jinping has taken a more cautious tone, insisting on the need to keep technology under human control. The choice of words thus reflects distinct political approaches, even though both leaders have participated in exchanges concerning the issues of AI.
A communication channel dedicated to AI-related incidents
At the end of the summit, China and the United States announced the creation of a communication channel concerning incidents associated with artificial intelligence. This initiative aims to provide a means of exchanging information when events arise that may raise concerns or require clarification between the two countries.
The existence of such a channel does not imply that Beijing and Washington share the same analysis of risks, nor that they have agreed on common rules applicable to all technologies. It does, however, indicate that both governments recognize the importance of maintaining a dialogue regarding AI-related incidents, in a context where systems are rapidly evolving and their use may have consequences beyond national borders.
Managing these issues raises practical challenges: determining what constitutes an incident, exchanging information quickly enough, and maintaining communication continuity when political relations are strained. The available announcements do not specify the detailed functioning of the mechanism nor the procedures that might be used. The channel thus constitutes, at this stage, a framework announced for exchanges rather than a complete agreement on AI governance.
A technological rivalry that also involves language
Terminology accompanies broader discussions on competitiveness, security, and control of advanced technologies. The United States and China are both seeking to influence the evolution of artificial intelligence, while their leaders must also address their public’s concerns regarding its economic and social effects.
The American choice to speak of “super-intelligence” can be read as a way to present the issue in a more ambitious light than that of a simply automated technology. However, this formulation does not necessarily align with the common scientific use of the word and may blur the distinction between current systems and a potential intelligence that exceeds human capabilities. The Chinese response, by remaining focused on respecting the American formulation, avoids publicly addressing this question.
Debates on AI also encompass the concentration of technological power, promises of innovation, and the conditions for its regulation. These tensions are notably addressed in an analysis dedicated to the paradox of Sam Altman between technological ambitions and political discourse, which illustrates the diversity of positions expressed around the development of artificial intelligence.
Different positions on technology control
The statements attributed to the two presidents reveal a difference in tone. Donald Trump has downplayed the fears of an AI threat to humanity several times, while Xi Jinping has emphasized the need to maintain human control. These positions do not alone describe the entirety of the policies of the two countries, but they show that their leaders highlight different priorities in the public debate.
The question of control encompasses multiple dimensions: the oversight of systems, responsibilities in case of damages, infrastructure security, and limits to impose on the most sensitive uses. It also depends on the governments’ ability to communicate about incidents and to prevent technical or political misunderstandings from escalating.
In this context, the exchange of statements surrounding “super-intelligence” is but one aspect of a broader relationship. The term chosen by Washington draws attention, but discussions on communication mechanisms and the maintenance of human control show that the concrete issues far exceed the question of the name given to technology.
