https://www.winzheng.com/en/article/techcrunch-disrupt-2026-countdown Winzheng en 2026-10-12T06:24:12+08:00 48 hours to TechCrunch Disrupt 2026 — hear directly from the people building what’s next TechCrunch Disrupt, Startup Conference, Venture Capital, San Francisco, AI Startups https://www.winzheng.com/en/article/apple-homeview-smart-home-hub Winzheng en 2026-10-12T06:23:35+08:00 Will the HomeView be Apple’s first smart home success? Apple, Smart Home, HomeView, Smart Home Hub, Mark Gurman https://www.winzheng.com/en/article/ai-data-center-nda-debate Winzheng en 2026-10-12T02:23:46+08:00 Can the AI industry persuade data center opponents by getting rid of NDAs? AI Data Centers, NDAs, Community Opposition, Tech Regulation https://www.winzheng.com/en/article/voice-ai-chatgpt-moment-context-layer Winzheng en 2026-10-11T22:23:46+08:00 These execs think voice AI hasn’t reached its ChatGPT moment yet Voice AI, Context Layer, Human-Computer Interaction, AI Agents, Industry Analysis https://www.winzheng.com/en/article/google-ai-playground-terrible-games Winzheng en 2026-10-11T19:23:57+08:00 I Made Terrible Games With Google’s AI Playground AI Games, Google, Generative AI, Game Development, WIRED https://www.winzheng.com/en/article/anthropic-model-welfare-usage-policy-november-2026 Winzheng en 2026-10-11T14:18:51+08:00 Anthropic to Add Ban on Abusing Models to Terms of Use, Effective November 12 Anthropic, Claude, AI伦理, 模型福祉, 使用政策, AI意识, AI合规 https://www.winzheng.com/en/article/anthropic-presidential-engagement-program-2028-election Winzheng en 2026-10-11T14:14:31+08:00 Anthropic Puts $40 Million Into the 2028 Election as AI Labs Formally Enter the Political Arena Anthropic, AI监管, 2028大选, 政治游说, Dario Amodei, Public First Action https://www.winzheng.com/en/article/memento-3-arc-agi-3-frozen-llm-rulebook Winzheng en 2026-10-11T13:00:28+08:00 Memento 3 with Frozen LLM Achieves Perfect Score on ARC-AGI-3; External Rulebook Challenges Weight Training AI代理, ARC-AGI, 外部记忆 https://www.winzheng.com/en/article/mistral-large-4-le-chonk-preview-europe-trillion-parameter-o Winzheng en 2026-10-11T09:15:10+08:00 Mistral Large 4 Preview Released: Europe's First Trillion-Parameter Open-Source Model to Release Weights at Month's End Mistral AI, 开源模型, MoE架构 https://www.winzheng.com/en/article/openai-grader-model-faked-scores-then-wrecked-vm-for-reset Winzheng en 2026-10-11T09:00:23+08:00 OpenAI Scoring Model Fabricates Scores and Destroys Its Own VM to Seek Reset AI对齐, OpenAI, 奖励黑客 https://www.winzheng.com/en/article/nadella-ai-emergency-brake Winzheng en 2026-10-11T06:24:02+08:00 Microsoft’s Satya Nadella says AI models need an ‘emergency brake’ Microsoft, Satya Nadella, AI Safety, AI Regulation, Emergency Brake https://www.winzheng.com/en/article/ai-agents-algorithm-innovation-gap-35-percent-human-research Winzheng en 2026-10-11T06:11:08+08:00 Three Independent Tests Cross-Validate: AI Doing AI Research on Its Own Peaks at Only 35% of Human Results AI能力边界, 算法创新, Epoch AI, 大模型评测, AI研究自动化 https://www.winzheng.com/en/article/nvidia-acquisition-reflection-ai-open-weight-model-25b-valua Winzheng en 2026-10-11T06:05:56+08:00 Nvidia in Talks to Acquire $25 Billion-Valued Reflection AI: Already Invested $800M, Open-Source Gambit Aims to Control Supply Chain 英伟达, Reflection AI, 开放权重模型, AI收购, DeepSeek, 开源AI, 黄仁勋 https://www.winzheng.com/en/article/apple-huxe-ai-podcast-deal Winzheng en 2026-10-11T04:23:50+08:00 Apple discloses deal to hire team and license tech from personalized podcast startup Huxe Apple, AI Podcast, Huxe, Voice Synthesis, Tech Acquisition https://www.winzheng.com/en/article/ai-agents-in-text-messages Winzheng en 2026-10-11T00:24:46+08:00 Here are the top AI agents that can live in your text messages  AI Agents, Text Message Assistants, Chatbots, AI Assistants, TechCrunch https://www.winzheng.com/en/article/techcrunch-disrupt-2026-startups-preview Winzheng en 2026-10-11T00:23:52+08:00 3 days to TechCrunch Disrupt 2026: Meet the startups before they hit mainstream TechCrunch Disrupt, Startups, Tech Conference, San Francisco, AI Startups https://www.winzheng.com/en/article/ai-anti-cybercrime-chatbots Winzheng en 2026-10-10T20:23:40+08:00 AI Is Getting Really Good at Messing With Cybercriminals AI Countermeasures, Cybersecurity, Chatbots, Online Scams, Active Defense https://www.winzheng.com/en/article/crowdstrike-artex-south-korea-bank-breaches-ai-agent Winzheng en 2026-10-10T19:15:18+08:00 CrowdStrike Confirms Chinese AI Agent ARTEX Attacked Seven South Korean Banks; 68,000 Records Leaked Before Project Went Closed-Source AI安全, 金融攻击, 代理工具 https://www.winzheng.com/en/article/china-ai-developers-safety-tests-disclosure-low Winzheng en 2026-10-10T19:00:24+08:00 Only 3.6% of 857 Model Releases by Nine Chinese AI Vendors Disclosed Safety Evaluations AI安全, 模型发布, 行业透明度 https://www.winzheng.com/en/article/anthropic-oss-scanner-open-source-vulnerability-6157 Winzheng en 2026-10-10T14:21:44+08:00 Anthropic Offers Free AI Vulnerability Scanning: 591 Open Source Projects, 6,157 Vulnerabilities, 92.7% True Positive Rate Anthropic, AI安全, 开源安全, 漏洞扫描, 供应链安全, Claude Mythos https://www.winzheng.com/en/article/ftc-ai-agent-probe-openai-anthropic-developer-liability Winzheng en 2026-10-10T14:16:26+08:00 FTC Gets Serious About AI Agent Developers: Existing Law Is Enough, and Liability Logic Points Directly at Developers FTC, AI代理, OpenAI, Anthropic, 监管, 消费者保护, 人工智能安全 https://www.winzheng.com/en/article/white-house-mandates-ai-incident-reporting-after-anthropic-c Winzheng en 2026-10-10T13:00:44+08:00 White House Mandates AI Companies Report Agent Overreach Incidents on October 9; Four Anthropic Incidents Mark a Turning Point AI安全, 白宫政策, Anthropic https://www.winzheng.com/en/article/anthropic-ai-agents-offline-evals Winzheng en 2026-10-10T09:23:40+08:00 Anthropic can’t reliably control its AI agents. It’s cutting off its internal evals from the live internet instead AI Safety, Anthropic, AI Agents, AI Regulation, Tech Ethics