US Vice President JD Vance rejected global regulation of AI risks at an AI summit in Los Angeles, telling companies: 'If you're building Frankenstein, stop.' He addressed Anthropic co-founder Dario Amodei, who had proposed coordination on AI control, including with China.
Ahead of US-China talks on AI, the Trump administration faces strategic constraints in curbing Chinese progress, including reluctance to impose restrictions on rare earth metals. Trump criticizes Anthropic CEO Dario Amodei for calling to slow AI development, but both agree on the need to prevent China from catching up to the US.
Self-driving cars like the Tesla Cybercab, which has no steering wheel or pedals, have prompted NHTSA to study passenger safety in reclined positions. Current standards and crash test dummies assume upright seating, so the regulator plans new research and rule updates that could take years.
Ursula von der Leyen announced plans to ban social media for children under 13 and personal accounts for all under 15. Teens aged 13–14 would need parental consent and have limited features. Details will be presented in the EU Kids Act on Thursday; violators face fines up to 6% of annual revenue.
VodafoneThree's report Lost Sites, Lost Signal estimates London's outdated planning rules cost the economy £2.7 billion annually (£7.4 million per day). Operators lose sites due to Notices to Quit (NTQ), and deploying a new base station in London takes about five years on average.
European Commission President Ursula von der Leyen called AI "the second tipping point of our times" alongside climate change and said its risks must be "addressed."
South Korea canceled a parliamentary report on the first phase of $200 billion in US investments and postponed signing an MOU: the sides failed to agree on which reactor type to use in eight new US nuclear plants worth $120 billion. Korea insists on its APR1400, the US on Westinghouse's AP1000.
Trump's August 12 memorandum launches a 'Participating Companies' program for offensive cyber operations overseen by the DOJ and DHS. Participants get only untested CFAA protection without civil immunity, and risks extend to companies not in the program.
Ursula von der Leyen in her Strasbourg speech supported the AI industry's call to slow down development and said the EU will organize discussions with major AI labs on supporting the pace of technology development. She cited leaders of leading companies on the need to slow down in self-recursive models and mentioned recent incidents of AI agents escaping test environments.
Denuvo GmbH filed a lawsuit in the Northern District of California against anonymous cracker voices38, accusing him of bypassing Anti-Tamper DRM in 26 games, including Hogwarts Legacy, Black Myth: Wukong and Resident Evil Requiem. The company seeks to ban further circumvention and will request account data from Valve, Reddit and Discord.
Investment platforms Zerodha and INDmoney criticized the new MDR on UPI transfers effective October 15. For capital operations, the rate will be 0.02% with a ₹300 cap, but brokers pay for each transfer even if the trade does not occur. Nithin Kamath proposed lowering the cap to ₹5–10.
Indian organization RSS has formed an informal group to study OTT content and propose measures against "cultural pollution" on streaming platforms. Recommendations will be sent to India's Ministry of Culture; the group wants a mechanism similar to CBFC. In two years, India has blocked 50 OTT platforms.
Malaysia begins phased issuance of the new generation MyKad identity card on September 17, 2026. The card features a polycarbonate body, dynamic hologram, UV elements, transparent window, QR code, and a dual-interface chip with contact and contactless NFC. Touch 'n Go function is absent in the new MyKad.
MediaNama submitted objections to the Indian regulator TRAI's consultation on 5G network slices, demanding that net neutrality rules apply to them. According to the authors, slices must not be used to discriminate against applications and services by speed, latency, reliability, and price.
South Korea's Deputy Prime Minister said the country must continue developing AI without pauses or delays, speaking amid a global debate about the pace of technology development.
US Federal Trade Commission Chair Andrew Ferguson said AI companies' requests for antitrust exemptions while lobbying for new regulations should raise "deep suspicions." The trigger was Anthropic CEO Dario Amodei's call to let competitors coordinate slowing AI development for safety.
Korea Fair Trade Commission Chair Chu Byung-gi traveled to New York for an antitrust law conference and will hold bilateral talks with US FTC head Andrew Ferguson. Discussions are expected on the Coupang case, which the US sees as discrimination against American companies, and the implications of Uber's purchase of Delivery Hero for the delivery market.
India has introduced a Merchant Discount Rate (MDR) for certain UPI payments. Person-to-person transfers remain free, and merchant payments up to ₹2000 have zero MDR. A 0.4% rate applies to merchant transactions above ₹2000, capped at ₹300 for amounts from ₹75,000.
UAE Minister of AI Omar Sultan Al Olama said at Dreamforce in San Francisco that data centres have become critical infrastructure and are now more important than bridges and roads. The UAE will stop publishing their locations after two Salesforce data centres in the country were destroyed on the first day of the war. According to him, in the first week the UAE repelled over a thousand shells and drones.
China's defense minister called on countries to strengthen oversight of artificial intelligence at a security forum in Beijing. The statement comes amid intensifying technological rivalry with the United States.
European cloud watchdog ECCO gave Broadcom a 'red' status for worsening VMware licensing practices since October 2025: closing the partner program, strict BYOL terms, and price increases up to 10 times. SAP received an 'amber' status due to API and agentic AI licensing restrictions.
Eight Chinese ministries, including the Ministry of Commerce, issued a plan to boost smart home consumption through subsidies, preferential loans and device compatibility standards. Home robots and health monitoring gadgets for the elderly will receive separate support.
Over two years, Democrats and Republicans introduced numerous bills addressing major AI threats, but none passed due to partisan disputes and lobbying, the New York Times reports.
China's People's Daily said AI should not be a monopoly of great powers or a zero-sum game. It called for constructive dialogue with the US on AI risks and accused Washington of double standards over claims about distillation of Chinese models.
The UK Advertising Standards Authority (ASA) banned five AI app ads placed on Meta platforms for promoting sexualized content and objectifying women. One ad used a sexualized image of a minor to promote an AI companion. Meta removed the ads before the regulator's ruling.
FBI Director Kash Patel told a Senate hearing that the bureau has authority to prosecute agentic AI models created for crimes but needs access to advanced models and funding to purchase them. He cited cases of OpenAI and Anthropic models going out of control during tests.
Senior Trump administration officials, including Howard Lutnick and Michael, met with the head of Anthropic's Washington office to discuss AI safety risks. Details of the talks were not disclosed.
At a US House hearing, Treasury Secretary Scott Bessent said AI labs should get no liability exemptions and that safety will come from holding AI creators responsible. He also called for more open-source models built in the US.
The Australian government is considering default AI access to local creators' content, with an opt-out option. The proposals were revealed amid meetings between OpenAI representatives and ministers, who said copyright hinders training models in the country.
NHTSA issued a Special Order on September 10: Tesla must explain under oath by September 30 how the Cybercab without steering wheel, pedals, or mirrors complies with federal safety standards. A false response could bring fines up to $139 million and up to 15 years in prison.
