European and Asian regulation, multilateral negotiations, and the cross-border movement of researchers and compute.
The first U.S.-China summit centered on artificial intelligence closed without safety commitments, as both governments signaled they intend to leave development untouched.
International correspondent · September 27, 2026The company said Tuesday's redesign — forced by a €460 million fine — elevates comparison platforms over local businesses and follows an earlier compliance change that Google says already cut free direct-booking referrals to European businesses by 30 percent.
Treasury Secretary Scott Bessent is expected to lead a mid-September bilateral dialogue on A.I.-directed cyberattacks and voluntary lab self-policing, days after both governments signed the Carolina Principles at the G20.
OpenAI's largest advertising expansion is scheduled to begin August 24, giving businesses in Germany, France, Spain, Italy and the Nordics access to sponsored placements inside a chatbot the company says has one billion weekly users.
The company said the expansion will begin next week, adding conversion bidding, geo-targeting and custom audiences to a platform where one in five weekly users, it says, is showing commercial intent.
Shares of the Hangzhou-based android maker closed 460 percent above their I.P.O. price on Wednesday, valuing the company at 342 billion yuan and raising $904 million in a landmark listing for China's embodied A.I. sector.
The iPhone maker becomes the first foreign firm cleared by Beijing to deploy a proprietary large language model on the mainland, ahead of an Apple Intelligence rollout expected in the coming months.
The European Commission's A.I. Office activated its investigative and enforcement powers over general-purpose model providers on Aug. 2, exposing OpenAI, Anthropic, Google and their peers to penalties of up to €15 million or 3 percent of worldwide turnover.
The European Commission's enforcement regime for general-purpose artificial intelligence took effect on August 2, drawing OpenAI, Anthropic and Google into direct regulatory scrutiny and adding a new front to transatlantic trade tensions.
A staff-led petition titled 'Pacing the Frontier' drew endorsements from OpenAI and Anthropic on Wednesday, two days before a Trump administration deadline under Executive Order 14409.
Michael Kratsios, the director of the Office of Science and Technology Policy, said the Beijing startup ran a 'sophisticated internal platform' to extract knowledge from Anthropic's Fable model; Treasury Secretary Scott Bessent said sanctions and Entity List designations are 'on the table.'
Moonshot AI's 2.7-trillion-parameter open-weight release closes the gap with Anthropic and OpenAI on coding benchmarks, unsettles markets, and reopens a policy fight the Trump administration had hoped to defer.
The World Artificial Intelligence Cooperation Organization, unveiled at the Shanghai conference and headquartered there, is being presented as an answer to the U.S.-led Pax Silica coalition of 35 countries.
A 2.8-trillion-parameter open-weight model from a Beijing startup, released as 29 nations signed the founding charter of a new intergovernmental A.I. organization headquartered in Shanghai, has upended the assumption that Chinese laboratories trail their American counterparts.
In his first address to the World Artificial Intelligence Conference, China's president announced WAICO, a 29-nation intergovernmental body, and pledged 5,000 training slots for the Global South.
The World Artificial Intelligence Cooperation Organization, headquartered in Shanghai and signed into being on July 16, gathers Russia, Brazil, Indonesia, Pakistan and two dozen other states outside the G7 and the E.U.
Speaking for the first time at China's World Artificial Intelligence Conference, President Xi Jinping formally launched WAICO, an intergovernmental body headquartered in Shanghai and operating outside the United Nations system.
The Chinese president will address the World Artificial Intelligence Conference in Shanghai on Friday — his first in-person appearance at the annual event — as Beijing moves to institutionalize a rival international A.I. governance order.
OpenRouter data show American developers routing as much as 46 percent of their token usage to Chinese systems. Hours later, Reuters reported that Beijing is meeting with Alibaba, ByteDance and Z.ai about restricting foreign access.
Secretary-General António Guterres told delegates that artificial intelligence is being deployed 'faster than anyone, including the people building it, can keep up,' as the inaugural Global Dialogue on A.I. Governance opened at the Palexpo on July 6.
At the inaugural Global Dialogue on A.I. Governance in Geneva, the Secretary-General warned that artificial intelligence is advancing 'at runaway speed' and unveiled a Child Safety Pledge — as his own 2026 deadline for a treaty on autonomous weapons passed without one.
Secretary-General António Guterres told delegates from 193 nations that an experiment was being run on the world's societies without a plan and without consent, and called for common safety baselines, human-rights red lines and a Global Fund for A.I.
Secretary-General António Guterres warned that artificial intelligence is 'being deployed faster than anyone, including the people building it, can keep up,' as an independent scientific panel delivered its first worldwide assessment of the technology's risks.
The two-day session at Palexpo opens Monday under the co-chairmanship of Estonia and El Salvador, and follows a preliminary scientific report warning that A.I. is advancing faster than governments can regulate it.
A 40-expert panel co-chaired by Yoshua Bengio and Maria Ressa delivered the first global scientific assessment of A.I. on Wednesday, days before member states convene in Geneva under the shadow of Washington's rejection of multilateral oversight.
In its first preliminary report, a 40-member independent panel convened by the General Assembly told member states that science cannot rule out catastrophic harm from advancing A.I. systems.
The Commerce Department reversed an 18-day directive that had forced Anthropic to disable its two most advanced models worldwide, ending a standoff with the Trump administration and raising unresolved questions about ad hoc government vetting of frontier A.I.
In a June 10 letter to the Senate Banking Committee, the company alleged that operators linked to Alibaba's Qwen lab used some 25,000 fraudulent accounts to conduct 28.8 million exchanges with Claude over six weeks.
In a June 10 letter to Senators Tim Scott and Elizabeth Warren, the A.I. lab said operators affiliated with Alibaba's Qwen program conducted 28.8 million exchanges with Claude through thousands of fraudulent accounts. Alibaba's Hong Kong shares fell to a 16-month low.
In a joint statement issued June 22, the cyber security leaders of the United States, United Kingdom, Canada, Australia, and New Zealand told boards and executives that the window to prepare for A.I.-accelerated attacks has narrowed to a matter of months.
The Commerce Department's June 12 directive barring foreign-national access to Fable 5 and Mythos 5 has forced a worldwide suspension that cybersecurity veterans now call dangerous to defenders.
The disclosure, made on the floor of the Senate by Mark Warner, reframes the Commerce Department's unprecedented order against Fable 5 and Mythos 5 as a question of autonomous offensive capability rather than a narrow jailbreak.
Two days after meeting Dario Amodei at the G7 summit in Évian-les-Bains, the president told Axios he no longer regards the company as a national security threat. The Commerce Department's June 12 order and the Pentagon's March supply-chain designation remain in force.
An export-control directive from the Commerce Department compelled Anthropic to disable Claude Fable 5 and Mythos 5 worldwide, setting a precedent for federal control of commercial frontier A.I.
The French president met Dario Amodei and Sam Altman in Évian to advance a 'trusted partners' scheme, after a Trump administration export-control order forced Anthropic to disable its Fable 5 and Mythos 5 models worldwide.
The Commerce Department's export-control directive, issued late Friday, forced the company to disable both models worldwide and has provoked alarm at the G7 summit and an open letter from cybersecurity researchers.
A Commerce Department directive barring foreign nationals from Fable 5 and Mythos 5 has prompted France to accelerate a pivot to domestic A.I. and pushed the question of American leverage to the center of the Evian summit.
The Japanese conglomerate will build 5 gigawatts of compute capacity across northern France by 2031, anchoring President Macron's Choose France summit and the European Union's sovereign-A.I. ambitions.
The 42,300-word document, signed on the 135th anniversary of Rerum Novarum, frames A.I. as a new industrial revolution and was presented at the Vatican alongside Anthropic co-founder Christopher Olah.
The Council and Parliament reached political agreement on May 7 on amendments to the bloc's flagship A.I. legislation, in a package the institutions have branded an 'omnibus' on simplification.