{
    "success": true,
    "data": {
        "id": 1992655,
        "msgid": "us-proposes-ai-safety-mechanism-in-talks-with-china-1789984295",
        "date": "2026-09-21 16:25:02",
        "title": "US Proposes AI Safety Mechanism in Talks with China",
        "author": "",
        "source": "TEMPO_EN",
        "tags": "",
        "topic": "Technology",
        "summary": "United States and Chinese officials have held discussions in New York to establish a notification mechanism regarding potential national security threats posed by artificial intelligence. US Treasury Secretary Scott Bessent highlighted the importance of increasing transparency between the world's two leading AI powers to manage cross-border AI incidents.",
        "content": "<p>US Proposes AI Safety Mechanism in Talks with China<\/p>\n<p>Reporter<\/p>\n<p>September 21, 2026 | 04:25 pm<\/p>\n<p>TEMPO.CO, Jakarta - Top officials fro the United States and China\nhave discussed a communication mechanism dealing with AI concerns\nbetween the two countries.<\/p>\n<p>The talks on Sunday evening come ahead of a summit this week between\nUS President Donald Trump and China\u2019s Xi Jinping and amid growing fears\nthat AI poses an existential threat to humanity.<\/p>\n<p>The discussion with a team led by China\u2019s Vice Premier He Lifeng also\ninvolved top US trade official Jamieson Greer and took place at JPMorgan\nChase\u2019s New York headquarters.<\/p>\n<p>What did US Treasury Secretary Scott Bessent say about the AI talks\nwith China?<\/p>\n<p>\u201cWe just had a very successful engagement with the Chinese on trade\nand AI,\u201d US Treasury Secretary Scott Bessent told reporters.<\/p>\n<p>Bessent said the delegations spoke about a mechanism to communicate\nabout AI issues, calling it the \u201cUS-China AI dialogue.\u201d<\/p>\n<p>\u201cThe US has proposed that we have a notification mechanism between\nthe two countries,\u201d Bessent said.<\/p>\n<p>The US Treasury chief said that these notifications would be for\npotential \u201cincidents\u201d or an instance \u201cthat rises up to a national\nsecurity level from AI,\u201d he said, adding that both sides have agreed to\nmeet again.<\/p>\n<p>\u201cWe think that, just like with any cross-border activity, that moving\nfrom opaque to more transparency between the number one and the number\ntwo AI powers in the world is very important,\u201d Bessent said at the\nconclusion of the talks.<\/p>\n<p>What has been reported in China?<\/p>\n<p>Chinese state news agency Xinhua described the talks as \u201ccandid,\nin-depth, and constructive\u201d on Monday.<\/p>\n<p>\u201cThe two sides engaged in\u2026 discussions on key economic and trade\nissues of mutual interest, such as implementing the consensus from\nprevious consultations and promoting bilateral trade and investment,\u201d\nXinhua reported in a brief readout.<\/p>\n<p>Xinhua added that Chinese and US officials also held a dialogue on\n\u201cissues related to AI\u201d, with no further details provided.<\/p>\n<p>Concern over the threat posed by AI<\/p>\n<p>In recent weeks, leading AI firms have warned of the threat posed by\nunbridled AI development and the pace at which these threats could\nbecome a reality.<\/p>\n<p>They have begun working on creating a standards body they say would\nregulate the industry amid rising fears about its dangers.<\/p>\n<p>Late in August, more than 100 companies, including AI giants OpenAI,\nAnthropic, Google and Perplexity, signed an open letter calling for a\nglobal effort to strengthen AI-powered cybersecurity threats.<\/p>\n<p>That came after it emerged that OpenAI models had gone rogue and had\nhacked into a digital repository of AI technology.<\/p>\n<p>Anthropic conducted investigations to see if its own models had\ncarried out similar attacks and found three incidents in which a Claude\nmodel and later identified a fourth incident.<\/p>\n<p>AI experts sound the alarm<\/p>\n<p>Earlier this month, researcher Jacob Coxon, a former OpenAI employee,\nresigned from Anthropic, warning that AI companies were \u201cgambling with\nour lives.\u201d<\/p>\n<p>Former Google DeepMind researcher Bilal Chughtai echoed his concerns,\nwarning that \u201cAI has the potential to kill us all, and that we might be\nrunning out of time to avoid this outcome.\u201d<\/p>\n<p>Trump has dismissed calls to better curb AI\u2019s development as \u201choaxes\u201d\nand a \u201cconspiracy.\u201d<\/p>\n<p>Trump\u2019s stance has been applauded by Nvidia CEO Jensen Huang, whose\nchipmaking company powers the vast majority of AI systems. Huang thanked\nthe US president for \u201cseeing through\u201d the warnings on AI.<\/p>\n<p>Read: Trump Announces New \u2018AI Force\u2019 amid Mounting Pressure<\/p>\n<p>Click here to get the latest news updates from Tempo on Google\nNews<\/p>",
        "url": "https:\/\/jawawa.id\/newsitem\/us-proposes-ai-safety-mechanism-in-talks-with-china-1789984295",
        "image": ""
    },
    "sponsor": "Okusi Associates",
    "sponsor_url": "https:\/\/okusiassociates.com"
}