U.S. proposes exchanging AI safety alerts with China, Bessent says, tell me about this news in (https://www.nbcnews.com/world/asia/us-proposes-exchanging-ai-safety-alerts-china-bessent-says-rcna598923)

U.S. Treasury Secretary Scott Bessent announced that the United States has proposed a reciprocal notification mechanism with China to share alerts regarding artificial intelligence incidents that could threaten national security.

Here are the key details of the development:

Proposed AI Safety Alert System

• Early Warning Mechanism: Under the U.S. proposal, Washington and Beijingthe world’s two leading artificial intelligence powerswould establish a direct communication channel to notify each other about major AI failures, rogue system behaviors, or AI-enabled cyber incidents that rise to the level of national security threats.

• Focus on Transparency: Bessent emphasized the need to transition from an opaque environment to greater transparency between the two nations, noting that clear communication is essential as AI capabilities advance rapidly and pose potential systemic risks.

Diplomatic and Economic Context

• High-Level Talks in New York: The proposal was presented during bilateral talks in New York between U.S. officialsled by Treasury Secretary Scott Bessent and U.S. Trade Representative Jamieson Greerand a Chinese delegation led by Vice Premier He Lifeng and chief trade negotiator Li Chenggang.

• Precursor to Leadership Summit: The discussions served as groundwork ahead of an upcoming summit between U.S. President Donald Trump and Chinese President Xi Jinping in Washington.

• Broader Trade Discussions: Alongside AI safety dialogues, the officials discussed operationalizing a bilateral “Board of Trade” to examine potential tariff relief on non-sensitive categories, including agricultural goods, energy products, and medical devices.

Responses and Challenges

• Chinese Reaction: Chinese state media confirmed that constructive discussions took place regarding economic, trade, and AI matters, though official readouts did not elaborate on the specific details of the U.S.-proposed alert mechanism.

• Underlying Tensions: While both nations recognize mutual vulnerabilities stemming from uncontrolled or catastrophic AI model failures, strategic competition remains fierceparticularly surrounding export controls on advanced semiconductor hardware and compute infrastructure.