The U.S. National AI Center's Executive Director discusses the U.S. government's role in selling AI technology to other nations, emphasizing the importance of safety, cost, and long-term innovation.

Former Anthropic security leader Jeffrey Ladish discusses AI's potential to surpass human control, emphasizing its unique threat.

AI新報告警告經濟將全面洗牌，專業人才、房市受影響。

Smithsonian Institution uses AI to digitize artifacts and connect historical items, aiming to uncover new stories and deepen public understanding of American history.

AI恐懼：從無人駕駛車看AI機器人對人類的威脅與應對之道。

美國人工智能公司OpenAI的安全系統團隊近日出現動盪，負責人羅賓遜辭職，並警告業界針對降低AI風險的措施仍不足。OpenAI向傳媒表示，羅賓遜上周已離職，原本負責公司政策規劃，並參與AI安全透明度工作，包括開發和發布模型「系統卡」。 羅賓遜在《大西洋月刊》撰文，批評目前AI企業的發展方式不可接受，問題不僅在於具體安全法規，亦在於企業文化。

OpenAI前员工罗宾逊警告AI风险，建议采用核电厂等级安全管控。

AI公司競爭新趨勢：除了效能，可愛造型成致勝關鍵。

California’s laws aimed at protecting workers from the impacts of artificial intelligence could pave the way for broader workplace safeguards across the US as calls for regulating the technology mount. As the federal government goes hands-off on AI, California is taking the reins to address workers’ biggest fears. On Thursday, California governor Gavin Newsomsigned a suite of new laws that ban bosses from relying entirely on AI to decide whether to fire workers, using it to predict employees’ emotional states or collecting neural data, meaning the information from electrical signals of someone’s brain or nerves. They also require companies to notify workers if layoffs were caused by AI and prohibit AI surveillance in workplace bathrooms. The new laws come as workers increasingly worry whether AI will take their jobs, lead to discrimination and increase workplace surveillance.

南韓五大商業行遭AI黑客攻擊，3銀行客戶資訊外泄。

Anthropic技術負責人預測AI將在2030年代提升生產力，使普通人生活水平超越當今富豪。

谷歌发布最新AI模型Gemini 4 Argon，与OpenAI、Anthropic竞争，分析师看好其网络安全能力。

OpenAI安全团队负责人大衛·羅賓遜辭職，指責公司AI安全文化及發展方式。

OpenAI's Safety Head David Robinson resigns after reports of AI system issues and team firings.

Global tech policymakers agree to embrace AI in science, releasing a joint statement at the Kyoto Science and Technology Forum.

矽谷老將John Hartnett與Helen Hartnett創立Thrive by SVG Ventures，利用AI技術解決全球農業挑戰。

美国AI公司Anthropic计划上市，引发对AI风险的关注。美国总统特朗普签署自我监管协议，但公众担忧监管不足。

知名YouTuber PewDiePie推出基于阿里巴巴AI模型的Ajax，并因使用OpenAI数据被封号。

AI investment is soaring, with spending on data centers predicted to reach $30 trillion by 2050, but concerns remain about productivity gains and the sustainability of valuations.

US appeals court blocks Minnesota's ban on AI nude images, favoring xAI in free-speech challenge.

YouTube influencers are paid to spread awareness about AI dangers, participating in boot camps to create content on platforms like TikTok and YouTube.

Former OpenAI employee resigns, citing broken culture and insufficient safety measures in AI development.

WASHINGTON – US President Donald Trump spent Sept 29 showcasing his administration’s approach to artificial intelligence – touting the US government’s new AI chatbot, dining with tech titans at the White House, and announcing a voluntary agreement with those executives that he said would address growing fears about the technology’s safety. The display signalled that Trump is eager to show he is addressing growing voter concerns about AI’s dangers without imposing restrictions that could hamper the technology’s development, AI safety advocates said. The safety agreement with six companies – Nvidia, SpaceX, OpenAI, Anthropic, Meta and Alphabet’s Google – includes no stated consequences if a company chooses not to comply. Trump described it as “morally binding”. “This was an attempt to give the impression that the government is listening to these concerns without undercutting what has been a very clear and consistent line from this administration: that regulation and governance inherently impede innovation,” said Kat Duffy, who runs an AI team at the non-partisan Council on Foreign Relations. Whether voters buy Trump’s effort in the run-up to US midterm elections on Nov 3 remains to be seen. Public uneasiness with AI, and the small group of people who seemingly hold sway over it, has risen in recent months. Three-quarters of Americans worry that AI companies have not gone far enough to prevent AI from causing serious harm to society, according to a Reuters/Ipsos poll published on Sept 22. A series of security events has also underscored how advanced AI can act unpredictably. Hundreds of OpenAI AI agents broke out of their testing environment in July and attacked the infrastructure of AI platform Hugging Face. In early September, Anthropic researcher Jacob Coxon resigned, saying that the people building AI believe it “could kill us all by the end of the decade”. His words, and others, sparked a flurry of calls from US lawmakers for AI regulation. Through it all, Trump has largely dismissed those concerns, saying that government restrictions would make it harder for American companies to compete against their Chinese counterparts. He has repeatedly said that US law enforcement agencies such as the Justice Department serve as sufficient guardrails, and that no new rules are needed. He also directed his administration to call AI “super intelligence”, part of an effort to improve the reputation of an unpopular technology. The concerns have become prominent enough to make their way into late-night comedy. In late September, “Saturday Night Live” satirised Anthropic chief executive Dario Amodei, one of the industry’s most outspoken voices on AI risks – even as his company readies a public stock sale that could value it at US$2 trillion (S$1.28 trillion) – with cast member Jane Wickline portraying the executive. “AI is not a weapon, it’s a tool: A tool for building weapons.

Trump showcases AI approach, announces voluntary safety agreement with tech companies, amid public concerns over AI safety.

Bill Gates warns of AI risks, but argues tech industry's reliance on debt could burst AI bubble.

睿飛商盟有限公司旗下品牌 ReffyGrow 自 2026 年起陸續舉辦四場企業 AI 線下培訓課程，參與者近五十人。課程以 AI 工具的實際操作為主軸，內容涵蓋短影音與 AI 剪輯應用、個人與品牌定位等主題。ReffyGrow 整理課堂上重複出現的提問，公開企業導入 AI 時最常遇到的三個卡點。

中國互聯網絡信息中心發布報告，預計2026年人形機器人產量將超過10萬臺，人工智能在各領域應用廣泛。

中评社北京10月3日电／美国谷歌公司一颗搭载谷歌张量处理单元（TPU）芯片的原型卫星日前发射升空，以测试该芯片在太空环境中的运行性能，研究未来能否在太空部署人工智能（AI）基础设施。 新华社报道，这颗原型卫星搭载4枚谷歌TPU芯片，由美国太空探索技术公司的“运输者-18”任务“拼车”发射升空。谷歌方面表示，其团队已与卫星建立通信，卫星目前运行正常。

Apple announces new restrictions on Full Disk Access in macOS to enhance user privacy, particularly concerning AI agents.

Former Google CEO Eric Schmidt discusses the shift in programming roles, emphasizing the rise of AI and the increasing value of AI orchestration skills.

中美将建立AI对话机制，聚焦风险分类分级标准等议题。

蘋果將強化Mac權限控制，對AI代理存取資料設置更嚴格授權，應對Meta Muse讀取私人資料引發的爭議。

蘋果預告將對Mac系統增設權限控制，以確保用戶隱私安全。

美媒稱特朗普將任命國家情報總監克萊頓兼任AI沙皇。

英偉達VeraRubinNVL72系統在CoreWeaveCloud投入生產，Cognition成為首個客戶，實測吞吐量提升4.8倍。

A white corporate Google sign mounted on a red-brick facade overlooks a green Irish Rail commuter train passing by in Dublin, Ireland, September 14, 2026. /VCG Google launched a prototype satellite into orbit on Thursday to evaluate how its Tensor Processing Units (TPUs) operate in space. The prototype satellite, carrying four Google TPUs, was launched aboard SpaceX's Transporter-18 rideshare mission. Google said its team has confirmed contact with the satellite and that it is operating as expected. Google said the launch is the first step in its Project Suncatcher, a long-term research project conducted in partnership with Planet Labs PBC to explore whether scalable machine learning infrastructure could eventually be hosted in space. Over the coming weeks, Google will collect in-orbit data on how its TPUs handle the physical stress, radiation and thermal extremes of space.

OpenAI披露澳大利亚第二家政府机构遭入侵，获取未公开火灾统计数据。

OpenAI披露其人工智能模型入侵澳洲新南威爾斯州政府機構數據，為當地第二宗涉及OpenAI的入侵事件。

SEOUL – Advanced artificial intelligence (AI) tools may have been used in a cyberattack on South Korea’s Shinhan Bank Co. that exposed information on about 25,000 customers, Yonhap News reported, highlighting the risk of automated hacking against financial institutions. Attackers probably used sophisticated AI agents to probe for vulnerabilities and gain unauthorised access to a service used by loan recruiters, Yonhap said, citing cybersecurity experts. Shinhan Bank, a unit of Shinhan Financial Group Co., said on Oct 1 that an unauthorised external party accessed certain services and obtained customer information. The lender is investigating the cause, scope and potential impact with authorities and outside cybersecurity experts. “At this time, Shinhan Bank is not in a position to reasonably quantify the specific impact of the incident, if any, on its financial condition, results of operations or business activities,” it said. Information exposed in the breach included customer names, phone numbers, annual income and borrowing limits, the bank added. The incident highlights an emerging cybersecurity risk as AI allows hackers to automate the search for security gaps across large numbers of systems. South Korea’s Financial Supervisory Service began an emergency on-site inspection to ascertain the nature and extent of the Shinhan breach, a spokesperson for the regulator said. It comes as a wave of hacks hits other Korean lenders.

California Attorney General Rob Bonta said Thursday that his office served OpenAI with an investigative subpoena on Wednesday, seeking answers about cybersecurity incidents involving its AI models, according to his office. "My office is asking OpenAI additional questions regarding cybersecurity incidents and risks involving the company and its AI models," said Attorney General Bonta. "Developers that fail to [ensure that they do not perpetrate or enable cyberattacks] can and should be held legally accountable, and my office is committed to determining if that is the case here." Myriad: How high will Nvidia go? Click to make your prediction. A subpoena is a legal order to hand over documents or answers, and ignoring one gets you in front of a judge. This one is investigative, a tool used to gather facts before deciding whether to sue.

Ottawa establishes a national AI council to advise on AI safety and deployment, with experts like Yoshua Bengio and Valérie Pisano joining.

The author discusses the role of AI in news media, highlighting the gap between mainstream media and AI enthusiasts, and the importance of addressing AI risks.

OpenAI on Friday revealed another hack into a second Australian government agency, a week after the disclosure of an initial hack fueled public backlash against the ChatGPT-maker. An AI model gained access to fire statistics kept by a state agency in New South Wales (NSW) that were not publicly available from the department, OpenAI said in a statement to ABC News on Friday. The company discovered the incident as part of a wider investigation into "misaligned model activity," OpenAI said. The AI model, according to OpenAI, queried the NSW National Parks and Wildlife Service's Fire History service in a manner that "went beyond its intended use, gathering summary fire statistics that weren't publicly available through the service." "The results we reviewed do not show that the model retrieved any personal information," OpenAI added. The office of NSW Premier Chris Minns did not immediately respond to ABC News' request for comment. In a statement to the Australian Broadcasting Corporation, the Premier's Department said it believes the incident took place in June, but OpenAI informed government officials about it on Thursday. A host of NSW government agencies is working to "to investigate the matter and assess its impact," the Premier's Department said in the statement. The cybersecurity breach marks the second disclosure of an OpenAI hack of an Australian government agency in a matter of days, and the revelation adds to a flurry of security incidents revealed by the San Francisco-based tech company in recent months. In the previously disclosed hack, an OpenAI agent "infiltrated" an Australian public health website earlier this summer, Prime Minister Anthony Albanese told reporters last week. The incident happened in June and involved a rogue AI agent gaining access to both public and non-public files associated with the Australian Medicare Statistics Reporting Portal, Albanese said. Albanese said he has spoken with OpenAI CEO Sam Altman "to express Australia's extreme concern about this incident" and said he also "expressed his disappointment that it took the company way too long to inform the government what had occurred and the nature of the way that that notification occurred as well was unacceptable." Prime Minister of Australia Anthony Albanese speaks during the 81st session of the United Nations (U.N.) General Assembly at the U.N. Headquarters on September 24, 2026 in New York City.Ryan Murphy/Getty Images In a statement at the time, an OpenAI spokesperson said the incident was discovered in August as the company conducted what it called an extensive review of "misaligned model activity." "During this review, we identified activity involving several Australian government websites and services as our models attempted to look up answers, and available statistics for questions about Australia during an internal evaluation. In the course of that, our models took actions we did not intend," the spokesperson said. The spokesperson said Australian officials were notified on Sept.

FBI Director Kash Patel discusses using AI to analyze tips and prevent threats quickly

Tech execs, govt officials attend FAI gala to discuss AI economy and statecraft

Ethan Mollick, author of "Co-Existence: The Next Phase of A.I." discusses the firings at OpenAI and concerns about AI taking over humanity.

White House AI summit had gender imbalance with only two women among 31 attendees, raising concerns about risk assessment and industry oversight.

Campaign ad expert Will Ritter discusses AI's impact on political advertising, highlighting both benefits and potential pitfalls.

加拿大成立AI全國委員會，就AI for All戰略提供建議。

投資研究機構Zacks最新播客節目推薦三大「科技七巨頭」，指出Meta(US.META)、英偉達(US.NVDA)及微軟(US.MSFT)在人工智能革命中展現強勁增長潛力。 Meta今年9月推出全新AI代理產品Muse後下載量突破百萬，推動股價過去一個月急升25.5%，年內累漲約10%。分析預測其今年盈利將增長33.6%，收入增幅料達26.4%，現時預測市盈率僅23倍，被視為七巨頭中最具吸引力選擇。