SIAgentNews.com
← All news

US and China to establish direct alert channel for AI incidents

Treasury Secretary Scott Bessent announced a new bilateral dialogue to manage risks of superintelligence while maintaining competition.

· First reported by The Straits Times

$30 billionnon-sensitive goods for mutual tariff cuts

The short version

  • A new Shenzhen dialogue will focus on "superintelligence" and sharing testing methods for advanced models.
  • The two nations agreed to a direct alert channel for serious AI incidents during a September White House meeting.
  • The US aims to "de-risk, not to decouple" from trade while pursuing a relationship based on balance and reciprocity.
  • A Board of Trade will oversee mutual tariff cuts on $30 billion of non-sensitive goods in each direction.

Why it matters

This signals a shift toward managed competition where the US and China agree to communicate on existential risks even as they vie for dominance. It establishes a framework for how the world's two largest economies will handle the safety of high-level AI.

A new framework for superintelligence

US Treasury Secretary Scott Bessent announced that the United States and China will engage in a new bilateral dialogue focused on "superintelligence" (SI). This initiative is set to take place during a meeting in Shenzhen on the sidelines of the Asia-Pacific Economic Cooperation leaders summit.

The primary goal of this dialogue is to manage the risks associated with advanced AI and prevent misunderstandings between the two nations' systems. Bessent emphasized that the two countries must remain clear-eyed about their differences while finding ways to cooperate on safety.

The dialogue will specifically look at exchanging best practices for evaluating and testing advanced models before they are released to the public. A key priority for both nations is ensuring that human judgment remains firmly in command of these powerful technologies.

Direct alerts and risk management

During a previous meeting at the White House in September, President Trump and Chinese President Xi agreed to create a dedicated line for notifying one another about major AI-related crises. This mechanism is designed to provide immediate communication if a significant technological crisis occurs.

Bessent noted that as superintelligence advances, the need for new areas of risk management will increase. The objective is to demonstrate how two nations can balance the demands of competition with the responsibilities of leadership in the AI space.

While the US seeks to maintain its leadership in the field, there is a recognized need to engage in these risk management efforts. The administration wants to reconcile the fact that national interests will not always coincide as the technology evolves.

De-risking without decoupling

The US approach to the Chinese economy is described as a desire to "de-risk, not to decouple." This means the US intends to manage specific risks associated with China rather than completely severing all economic and trade ties. Bessent framed this as a pursuit of a constructive relationship based on balance and reciprocity.

To support this, the two nations have agreed to several trade initiatives. These include the creation of a "Board of Trade" to decide on shared tariff reductions for $30 billion of non-sensitive items moving in both directions. Additionally, China has pledged to purchase more US farm goods.

A "Board of Investment" will also be created to provide a structured forum for market access. This board is intended to provide increased predictability for Chinese companies while addressing the friction caused by the ongoing US-China trade deficit.

Corporate and geopolitical friction

While the government focuses on high-level policy, the private sector faces its own scrutiny. Nvidia has faced questions regarding the remarks made by Bessent, as they did not address a controversy involving the company's chief executive. The CEO is currently under internal review following a report about a personal relationship with a woman suspected by US intelligence of being a Chinese asset.

The Chinese foreign ministry has dismissed these spying claims as a smear and entirely without basis. Meanwhile, Nvidia's board stated they have no evidence that any company secrets or national security details were leaked. The CEO has admitted to being unfaithful to his wife but has not shared any specific details about the situation.

These tensions highlight the complex overlap between personal conduct, corporate security, and international relations. As the US and China navigate the rules for superintelligence, the security of the companies building these tools remains a critical point of observation for both governments.

The path toward 2026

Bessent will travel to China next week for a meeting that precedes President Trump's third meeting with President Xi in 2026. These high-level interactions are intended to build a foundation for long-term cooperation on both economic and technological fronts.

The current strategy relies on recognizing where common purposes can prevail over competing interests. By focusing on specific areas like trade and AI safety, the US hopes to stabilize the relationship despite deep-seated differences in their approaches to every challenge.

The upcoming meetings serve as a testing ground for these new diplomatic tools. They aim to create a predictable environment where the two superpowers can coexist while navigating the rapid development of superintelligence. A fourth summit between the two leaders is expected to occur in December at a G20 gathering in Miami.

]}],

Sources and links

First reported by The Straits Times. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

More AI news