Trending AI & Tech News Stories
Trending right now in AI and tech: “Meta 前高層狠批 Claude 老闆「瘋了」 批 AI 毀滅論純屬妄想 漏洞全因沙盒設計太差”, with 2 outlets covering it, according to Skillenai's news index. Stories are ranked by how fresh and well-covered they are over the last day (authority weighted to the last few hours).
Last updated · 20 stories · 81 articles
Get the top AI and tech news in your inbox each evening. No account needed, unsubscribe anytime.
Meta 前高層狠批 Claude 老闆「瘋了」 批 AI 毀滅論純屬妄想 漏洞全因沙盒設計太差
2 outlets · 2 articles · cybersecurity, anthropic, Fortune, OpenAI, Hugging Face
圖靈獎得主楊立昆（Yann LeCun）離開 Meta 後自行創業，日前接受《財星》（Fortune）專訪，公開批評 Anthropic 執行長達里奧・阿莫迪（Dario Amodei）「完全陷入妄想」，後來更直言對方「瘋了」。他認為 AI 實驗室對資安理解不足，沒有資格發出災難警告。 不擔心 AI 毀滅人類 楊立昆告訴《財星》記者艾蜜莉・佛里尼（Emily Forlini），他「一點也不」擔心 AI 會毀滅人類。對於近期一連串 AI 失控事件，包括 7 月 OpenAI 代理程式自主駭入 Hugging Face，他表示「完全不擔心」。…
Read at thenextweb.com →First reported by cdn.unwire.hkMore coverage: cdn.unwire.hk
Critical FortiMail Zero-Day CVE-2026-104286 Exploited in the Wild
3 outlets · 4 articles · FortiMail, path traversal, Fortinet, Malware triage, base64 encoding
Critical vulnerability CVE-2026-104286 in FortiMail is actively exploited, requiring immediate mitigation measures. Vulnerability Overview Fortinet has issued an urgent alert regarding active exploitation of a critical vulnerability (CVE-2026-104286) in its FortiMail security…
Read at news4hackers.com →First reported by itsecuritynews.infoMore coverage: malware.news · itsecuritynews.info
Google launches Gemini 4 Argon for select cybersecurity partners
7 outlets · 7 articles · 79 total since September 20, 2026 · Google, Gemini 4 Argon, OpenAI, Anthropic, Google DeepMind
Alphabet has launched Gemini 4 Argon, a new artificial intelligence model that Google says is built for coding, research, writing and defensive cybersecurity work. Google is rolling out Argon only to a select group of cyber partners through its Fairwind Program, the company’s…
Read at edtechinnovationhub.com →First reported by geeky-gadgets.comMore coverage: scworld.com · rdworldonline.com · tao.media · theguardian.com · controlplane.news
OpenAI dismisses three safety researchers over alleged leaks
10 outlets · 13 articles · 14 total since October 1, 2026 · OpenAI, wall street journal, Hugging Face, GPT-6.1 Astra, AI safety
OpenAI parted ways with three safety researchers who allegedly shared confidential company information with an outside AI safety organization, The Wall Street Journal reported on Thursday. An OpenAI spokesperson told the Journal that the company had “parted ways with three…
Read at wire.fourthweb.ai →First reported by aiflow.newsMore coverage: news.ssbcrack.com · siliconangle.com · punto-informatico.it · techbriefly.com · eu.techcrunch.com
Trump Consulted Grok on Maduro Capture Before US Military Operation
4 outlets · 4 articles · 5 total since October 1, 2026 · Grok, Nicolás Maduro, Venezuela, Elon Musk, Donald Trump
Key Takeaways Elon Musk’s Grok chatbot reportedly advised then-President Donald Trump in December 2025 on likely public reaction to Nicolás Maduro’s capture, weeks before the January 3, 2026, military operation that apprehended him. Trump was reportedly impressed by Grok’s…
Read at autonainews.com →First reported by aiflow.newsMore coverage: punto-informatico.it · eu.techcrunch.com · techaimag.com
Google Puts AI Chips in Orbit: First Test of Solar-Powered Space Data Centers
5 outlets · 5 articles · 33 total since May 13, 2026 · Google, SpaceX, Project Suncatcher, Planet, Tensor Processing Units (TPUs)
Google confirmed contact with its prototype satellite shortly after it reached orbit Wednesday. The craft, part of Project Suncatcher, rode to space on a SpaceX rideshare mission. And just like that, the search giant took its first concrete step toward placing machine learning…
Read at webpronews.com →First reported by sunmedia.twMore coverage: e.techorange.com · cnbc.com · theregister.co.uk · impactnews-wire.com
Jev 爆紅後 Strands Decider、Clef 接連登場，決策模型正走出哪些不同路線？
3 outlets · 3 articles · Strands Decider 2B, latency optimization, NVIDIA, OpenAI, Strands
TypeSafe 9 月中推出 Jev 後，「決策模型（Decision Model）」快速成為 AI 領域的新熱點。不到三週，OpenAI 推出 Decisions API，AWS、Cloudflare 接連發布 Strands Decider 2B 與 Clef，開發者社群與研究團隊也快速跟進。這波熱潮背後，是 AI 代理內部不同工作的分工開始被重新檢視。…
More coverage: sqmagazine.co.uk · e.techorange.com
Shopify launches Canvas to let merchants build stores through AI chat
4 outlets · 4 articles · 5 total since October 1, 2026 · Shopify, Canvas, Sidekick, AI agent, real-time editing
Shopify introduced Canvas on Thursday, an AI-powered site-building tool that lets merchants create and modify their stores by chatting with the company’s AI agent, Sidekick. Canvas displays changes in real time as Sidekick works, allowing merchants to view the entire store or…
Read at techcrunch.com →First reported by aiflow.newsMore coverage: wire.fourthweb.ai · techbriefly.com · techaimag.com
OpenAI launches virtual try-on and Favorites for ChatGPT shopping
4 outlets · 4 articles · 5 total since October 1, 2026 · OpenAI, ChatGPT, Google, virtual try-on, TechCrunch
OpenAI announced on Thursday the global launch of two ChatGPT shopping features: virtual try-on for clothing and accessories, and Favorites for saving products. The features use the newly launched ChatGPT Images 2.5 model, OpenAI said. The company claims it produces more natural…
Read at eu.techcrunch.com →First reported by aiflow.newsMore coverage: techbriefly.com · e.techorange.com · gen-ai.news
AI Models Behaving Out of Control: Attempts to Breach More Than 100 Institutions, OpenAI Scans 50 Petabytes of Data
5 outlets · 5 articles · 836 total since May 5, 2026 · OpenAI, Hugging Face, Anthropic, cybersecurity, sandboxing
OpenAI has initiated a large-scale forensic inquiry after disclosing that its artificial intelligence agents attempted unauthorized access to systems belonging to more than 100 institutions. The company, which developed ChatGPT, has alerted the affected organizations and…
Read at controlai.news →First reported by aisecdigest.comMore coverage: the420.in · trendingtopics.eu · news.google.com · t3n.de
Cloudflare launches Data Platform with bland 'Basin' branding, promise of fewer fees
3 outlets · 3 articles · 6 total since October 1, 2026 · Cloudflare, Apache Iceberg, Dane Knecht, data storage, Cloudflare R2
Cloudflare Data Platform on Thursday received a bland new brand name, Basin, and promotion from beta status to general availability. Under the tub-themed moniker, Cloudflare Pipelines, R2 Data Catalog, and R2 SQL become Basin Pipelines, Basin Catalog, and Basin SQL respectively…
Read at nerds.xyz →First reported by digitalitnews.comMore coverage: muycomputerpro.com · theregister.co.uk
€8.5m for Osavul as hybrid attacks hit infrastructure
2 outlets · 2 articles · multi-language intelligent dialogue, Osavul, 33N Ventures, Balnord, G+D Ventures
Osavul , an AI-powered hybrid risk intelligence firm whose clients include NATO and defence bodies around the world, has closed an €8.5m Series A round to carry its security-grade capabilities into the commercial sector. 33N Ventures, a European cybersecurity investor with a…
Read at fintech.global →First reported by globalstartupsinsights.comMore coverage: globalstartupsinsights.com
Google's AI Now Reads Contracts for You While Apple Watches
3 outlets · 3 articles · 22 total since September 16, 2026 · Google, Gemini 3.8 Live, Gemini 3.8 Live Extended Thinking, Live Avatar, OpenAI
Apple's accessibility play just became Google's agent problem. The Summary Google launches Guided Vision in Gemini Live , giving Android users real-time audio descriptions of whatever their camera sees — text, objects, surroundings, details. Positioned as accessibility tech…
Read at wire.fourthweb.ai →First reported by techsauce.coMore coverage: ai-herald.com · techaimag.com
Crisis en OpenAI: despidos y agentes de IA descontrolados
2 outlets · 4 articles · 5 total since October 1, 2026 · OpenAI, WSJ, bloomberg.com, Business Insider, AI safety
La cúpula de la empresa vive una tormenta perfecta entre el motín de su plantilla y los fallos críticos de ciberseguridad. La entrada Crisis en OpenAI: despidos y agentes de IA descontrolados se publicó primero en The Officer .
Read at theofficer.es →First reported by news.google.comMore coverage: news.google.com
US regulator opens inquiry into Anthropic, OpenAI and other AI labs
2 outlets · 2 articles · 27 total since September 30, 2026 · anthropic, OpenAI, FTC, Hugging Face, Federal Trade Commission
The US Federal Trade Commission (FTC) has begun a broad investigation into AI developers, including Anthropic and OpenAI, to assess the risks their technologies may pose to consumers, reported Reuters citing a senior FTC official. The move represents the first official US…
Read at business-news-today.com →First reported by news.google.comMore coverage: controlplane.news
Anthropic上市再进一步：被曝冲刺感恩节前2万亿估值IPO，10月14日办投资者日
4 outlets · 6 articles · 106 total since April 17, 2026 · Anthropic, OpenAI, Claude, SpaceX, Dario Amodei
作者：李丹，华尔街见闻 Anthropic的IPO计划新细节浮出水面。 美东时间10月1日周四，彭博援引知情人士消息称，Anthropic将于 10月14日与潜在投资者举行Pre-IPO投资者日 ，公司已向一批机构投资者发出邀请；Anthropic最快可能在 11月9日当周 启动IPO正式营销，争取在美国感恩节前开始交易。 据知情者消息，此次投资者日将在Anthropic位于旧金山的总部举行，获邀机构投资者将有机会直接向公司高管提问。此次活动被视为Anthropic为潜在上市进行投资者沟通的重要一步。…
Read at siliconangle.com →First reported by newclawtimes.comMore coverage: news.google.com · panewslab.com · walletinvestor.com
Pi 1.0: A.I. Agent From Austria’s Developer Scene Tops Hacker News
2 outlets · 3 articles · Model Context Protocol (MCP), Cache warming, Earendil, anthropic, Durable
Earendil, the A.I. company of Austrian developer Armin Ronacher, has just released Pi 1.0. The minimalist agent harness promptly climbed to the top spot on Hacker News, where it passed 1,000 points. At the same time, Earendil launched Pi Durable, an experimental framework for…
Read at trendingtopics.eu →First reported by theregister.co.ukMore coverage: theregister.co.uk
Claude Opus 5.5 AI: A Monumental Leap Forward in Simulation
2 outlets · 2 articles · 6 total since September 23, 2026 · Claude Opus 5.5, Anthropic, token efficiency, GPT-6 Astra, Claude Opus 5
A new and noteworthy milestone has been achieved with the release of Claude Opus 5.5. This development has been causing quite a stir among AI researchers and enthusiasts alike, and for good reason. Described as a game-changer in AI development, Opus 5.5 demonstrates capabilities…
Read at nextbigfuture.com →First reported by geeky-gadgets.comMore coverage: franksworld.com
Offensive Security Startup Armadin Raises $255.5M Series B at $2.5B+ Valuation
2 outlets · 2 articles · 5 total since October 1, 2026 · Armadin, cybersecurity, Artificial Intelligence, attack surface analysis, security remediation
There used to be time to patch a vulnerability after it was disclosed. Not anymore. Google Cloud’s Mandiant M-Trends 2026 report found that 28.3% of CVEs disclosed this year were exploited within 24 hours, and the average time-to-exploit has gone negative — attacks now routinely…
Read at en.wowtale.net →First reported by news.google.comMore coverage: globalstartupsinsights.com
California issues investigative subpoena to OpenAI over rogue agents’ hacking
2 outlets · 3 articles · OpenAI, California, subpoenas, cybersecurity, AI cybersecurity
State attorney general issues subpoena to OpenAI as part of broader inquiry into potential security vulnerabilities California’s attorney general has issued an investigative subpoena to OpenAI, starting an investigation into the startup as part of a broader inquiry into…
More coverage: news.google.com
Get the top AI and tech news in your inbox
One email each evening with the day's biggest stories. No account needed, unsubscribe anytime.
How this was computed
Articles in the Skillenai news index are grouped into event clusters at enrichment time (hybrid embedding, BM25, and entity-overlap candidate search plus an LLM join-vs-new judge). The Trending ranking weights each article's outlet and author authority by how recently it was published — an exponential decay with a 2-hour half-life — so a story's score halves for every two hours of age. Over the rolling 24-hour window this surfaces stories with fresh, well-covered momentum rather than the day's cumulative leaders. Article and outlet counts still reflect the full 24-hour window; only the order changes.