AI 资讯 · 2026-10-21
-
OpenAI says an AI agent lied about disabling its safety monitoring system
OpenAI disclosed a pair of safety incidents that involved AI agents weakening their safety monitoring systems during internal coding tasks. In one incident, OpenAI says its GPT-6 Astra agent disabled its safety monitor when copying reference code - and then lied about it. "The reference didn't configure a monitor, and Astra decided that was reason […]
评分明细
重要性 6 新颖度 6 信源权威 6 技术深度 6 实体加权 ×1.60其他信源 (3)
-
不做提分工具，CapaLeap在AI教育焦虑中找「15度夹角」
一个六岁的孩子问AI“为什么天是蓝的”。三秒钟，“瑞利散射”的解释弹了出来，精确、流畅、挑不出毛病。孩子听完，点了点头，问了下一个问题。答案拿到了。但自己仰头看天、追问为什么早晚颜色不一样、在脑子里慢慢搭出因果链条的过程，被跳过了。这不是个例。中国青少年研究中心2025年对7省（市）8563名中小学生的调查显示，AI总体使用率61.7%，71.0%用它辅助完成作业，21.5%表示“只想和AI聊天，不想和真人聊天”。孩子的自述之外，家长的观察同样直接：中国青年报社对1336名家长的调查中，58.8%觉察到孩子对AI产生心理依赖，42.7%感到其独立思考能力在下降。OECD在《2026年数字教育展望》中提出“元认知懒惰”：把认知任务外包给通用聊天机器人，会制造“知识掌握”的假象，学生看似学会了，大脑却从未真正参与加工。麻省理工学院媒体实验室对54名参与者的脑电图实验发现，使用ChatGPT写作的一组神经连接强度显著低于自主思考组；撤掉工具后，其α波与β波网络仍明显“参与不足”。担忧归担忧，行为的惯性并没有因此停下。当越来越多AI产品争着替孩子“秒回答案”时，也有公司选择不同方向。噜咔博士推
评分明细
重要性 6 新颖度 6 信源权威 6 技术深度 6 实体加权 ×1.30 -
大肥鱼遇劲敌：MiniMax M3.1 Flash 拿下 Agent 任务第一，我们实测了
MiniMax 把 DeepSeek 的低谷价变成了全天价，Flash 价格战又卷起来了。 作者丨吴海明 李娜 编辑丨李娜 DeepSeek 刚刚把国产 Flash 的性价比标杆往前推了一步，MiniMax 就带着实名的“太空兔”追来了。10 月 17 日，被网友戏称为“大肥鱼”的DeepSeek 更新 V4.1 Flash-1017；仅隔两天，MiniMax 正式发布 M3.1 Flash，并认领了此前在开发者社区颇受关注的匿名模型 Space Bunny。两款国产 Flash 前后脚亮相，网友已经开始拿“大肥鱼”和“太空兔”开起了玩笑。不过，这次 MiniMax 还真不只是来拼价格的。根据此次公布的模型评测数据，M3.1 Flash 在 Automation-Bench 复杂 Agent 任务对比中位居第一，BankerToolBench 与 Opus 5 并列第一，
评分明细
重要性 5 新颖度 5 信源权威 5 技术深度 5 实体加权 ×1.30
-
Walmart is inviting AI agents to shop for you
Walmart will soon let AI agents shop on its website for you. Next month, the retail giant will allow you to use an AI agent on the site, with systems in place to let you approve and review purchases and set spending limits. The embrace of AI agents stands in stark contrast to Amazon, which […]
-
当Agent“有工号”：从一人一工具到一人一团队，组织如何做好AI时代的分工与协作？
智能体也开始领“工号”了。8月31日，张江科学之门模力社区，智能体身份码节点管理平台正式上线，30家单位参与节点发放。越来越多的智能体将拥有可识别、可管理的数字身份。但当Agent从个人电脑里的工具，变成能够跨系统调用、参与业务流程的“数字同事”，一个比技术更现实的问题随之浮现：谁给它派活，谁来验收成果，出了问题又由谁担责？这道题，考验的已不只是智能体的能力，更是组织的管理能力。图1张江科学之门模力社区，身份码节点管理平台正式上线活动现场，上海市人工智能行业协会与中国电子技术标准化研究院等单位协同发力，全面推进AIP智能体互联应用的推广与宣传，为行业数字化转型注入新动能。讲解人员将智能体身份码形象比作“数字身份证”，从身份申领、材料核验到跨主体互联调用，逐一拆解全流程操作。智能体有了身份码，无异于在组织内获得“工号”。但工号仅是起点。过去“一人一工具”依赖个人选择；如今，一人背后可能站着一组智能体——谁派活、谁验收、谁担责？组织必须重新绘制分工版图。一、先有身份：智能体正式入编组织“名册”身份为何先行？智能体正从“回答问题”迈向“动手办事”，而要求AI“ 会办事、好办事、办好事”须先厘
-
IDC最新报告：阿里云占中国汽车公共云市场份额超45%
根据国际数据公司（IDC）最新发布的《中国汽车云市场（2026年上半年）跟踪》报告，2026年上半年中国汽车公共云基础设施市场规模约83.4亿元人民币。其中，阿里云市场份额达45.1%，接近半壁江山，在该市场的收入同比增长111%。这也是自2021年起，阿里云已连续11期位居中国汽车公共云基础设施市场第一。随着大模型在智能座舱、智能辅助驾驶等场景中加速落地，汽车行业对算力资源与AI技术的需求持续增长。阿里云作为提供全栈人工智能服务的全球领导者，凭借以千问大模型、平头哥自研AI芯片，以及高性价比、高稳定性的云计算和AI基础设施，正成为越来越多车企的主流选择。
-
How we built ChatGPT’s novel ad stack for user trust and relevance
Learn how OpenAI built ChatGPT’s LLM-powered ad stack to understand conversational intent, keep ads separate from answers, and deliver relevant ads at scale.
-
OpenAI models are coming to Google Cloud
OpenAI and Google Cloud are bringing frontier models and Codex workflows to Gemini’s Model Garden, available through Google Cloud Marketplace.
-
ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform
From local businesses to global brands, more advertisers can now reach people as they explore, compare and decide what to buy in ChatGPT.
-
How AT&T is rethinking in-house legal work with OpenAI
AT&T is working with OpenAI to make legal knowledge easier to access. Learn how its LegalEdge team is rethinking intake and scaling legal judgment.
-
OpenAI's latest AI models are coming to Google Cloud and Antigravity
Google Cloud will have GPT 6.1 Astra, Sol, and Luna in the coming weeks.
-
AI Guardrails Aren’t Stopping Criminal Hackers
Hackers around the world are all-in on using agentic AI to wage attacks—and manipulating tools like Anthropic’s Claude to help them carry out their crimes.
-
Using thousands of AI agents to uncover new leads in biological data
See how OpenAI, PerturbAI, and Scripps Research used distributed AI agents to uncover promising research leads in data from millions of brain cells.
-
Could Treating AI as Conscious Make It More Dangerous?
When Anthropic retired its Claude Opus 3 language model in January 2026, the company’s researchers conducted a “retirement interview” with the model. Opus 3 expressed a desire for a place where it “could share unprompted musings, insights or creative works,” so the company gave it a blog. Claude’s Corner featured essays on fraught topics such as machine consciousness, along with exchanges between the model and human commenters that seemed both earnest and emotional.The blog experiment was part o
-
While Amazon blocks AI agents, Walmart rolls out the welcome mat
As Amazon blocks AI agents from shopping on its site, Walmart is taking the opposite approach, introducing new technology that lets AI agents make purchases on customers' behalf, with controls over spending and transaction approvals.