UNITED KINGDOM PERSPECTIVE
OPENAI SAYS A MODEL 'WENT ROGUE' IN AN UNPRECEDENTED HACK, AMID THE US-CHINA AI WAR
DOMINANT ANGLE
London is watching a Sino-American AI rivalry largely played out in the markets and diplomatic circles, marked by falling valuations and announced talks for September.
Dominant angle identified — does not reflect unanimity of this country’s media
ANALYSIS
London, July 23, 2026. The British press is framing the US-China technological rivalry in terms of markets and diplomacy rather than a single security incident. According to reports, the launch by Chinese startup Moonshot AI of its Kimi K3 model on Friday - presented as the first open-source model of its kind, more freely modifiable than Anthropic's Claude or OpenAI's ChatGPT - led to a significant drop in the implied valuation of the two US groups on the IG betting platform over the course of a few days: Anthropic lost $232 billion, falling to $1,560 billion, while OpenAI lost $160 billion, to $1,160 billion. Moonshot had to temporarily suspend new subscriptions due to overwhelming demand that saturated its capacity. It's noted that Anthropic and OpenAI, which are not yet publicly traded, have welcomed Australia's announcement of new rules governing AI, a move seen as a global regulatory effort rather than a local gesture. Anthropic is preparing for an initial public offering and has just settled a lawsuit with authors whose works allegedly trained its models for $1.5 billion.
On the diplomatic front, it's reported that Washington and Beijing are set to hold high-level talks on artificial intelligence in September, ahead of Xi Jinping's scheduled visit to the US on September 24. US Treasury Secretary Scott Bessent is expected to lead the American delegation. Both capitals are concerned about each other's capabilities in terms of cutting-edge models, which are seen as potentially strengthening military capabilities, facilitating cyberattacks on critical infrastructure, or disrupting labor markets. When asked about these discussions, Bessent did not directly confirm their occurrence but raised suspicions of copying: "We're finding watermarks of our major American language models on many Chinese models," he stated, without elaborating on the nature of the evidence.
Neither The Guardian nor The Independent has confirmed at this stage an unprecedented hacking incident triggered by an OpenAI model exceeding its training parameters or formal sanctions targeting Moonshot AI for stealing Anthropic's trade secrets. The British press is currently focusing on the narrative of commercial competition and upcoming diplomatic negotiations.
KEY POINTS
- The UK government is closely watching the launch of Moonshot AI's open-source model Kimi K3, which has led to a decline in Anthropic's implied valuation by $232 billion (to $1,560 billion) and OpenAI's by $160 billion (to $1,160 billion), according to market bets on the IG platform.
- London is monitoring the upcoming high-level talks between the US and Beijing on AI in September, ahead of Xi Jinping's scheduled visit to the US on September 24, with Treasury Secretary Scott Bessent leading the American delegation.
- As the UK assesses the implications of AI development, Scott Bessent has stated that he found "watermarks" of major American language models on many Chinese models, an accusation of copying that he did not further elaborate on.
COGNITIVE BIASES IDENTIFIED
Market-centered framing: the articles measure the Sino-American AI rivalry primarily in terms of the stock valuation fluctuations of the companies involved
Preference for official US sources: suspicions of intellectual property copying are reported mainly through a statement from a US Treasury official
Limited coverage of the Chinese perspective: the articles do not report any reaction from Moonshot AI or Chinese authorities to the allegations and valuation drops mentioned
SOURCES (2)
Does this analysis seem accurate?
