Pai 的 AI 雷达

今日必看

今天的简报围绕“代理正在如何工作、被评估和治理”展开：从 agent-sessions、前沿模型的 AEO 选择与应对，到数学代理作弊、Agent 线上 Badcase 和不确定性闸门，再延伸至 ChatGPT 与 Gemini 的流量变化、内容产品与独立新闻支持；同时夹带视频压缩、地图投影、洛杉矶城市演化和 MovieMaker 等可直接动手或轻松浏览的技术小品。阅读时可先把前半部分当作一组关于 Agent 可靠性、分发与社会影响的连贯线索，再用后半部分的工具、可视化和旧邮件案例换气，最后回到“如何在不确定中使用 Agent”这一实践问题。

日期
2026/09/08
今日条目
108
往期2026/09/08
全部108编码代理7Agent与工具23个人助理1模型与评测56创作成片2开发与开源1产品接入18
领域机器人具身1硬件边缘2生信医疗8游戏1安全治理4行业观察17

更多资讯

#29

OpenAI reports AI "research interns" and warns about its own pace at the same time

OpenAI reports that AI agents in its own research already handle 3.1 workdays for every human workday, and it says it has reached its goal of an "automated research intern." But chief scientist Pachocki warns that no lab has a good enough grip on alignment and monitoring to keep

推荐理由OpenAI reports AI "research interns"：OpenAI reports that AI agents in its own research already handle 3.1 workdays for every。
Agent与工具安全治理The Decoder
the-decoder.com打开原文
#39

Anthropic 据报道签约高达 5170 亿美元算力协议，锁定至少 14.8 GW 算力

据 The Information 报道，Anthropic 在十一个月内签署了价值高达 5170 亿美元的算力合同，自 2025 年 10 月以来锁定至少 14.8 GW 算力，并计划自建数据中心。 🔗 阅读原文 via AIHOT · https://aihot.virxact.com/items/cmtrkorpo07jdrotnozlrlyx2

推荐理由据 The Information 报道，Anthropic 在十一个月内签署了价值高达 5170 亿美元的算力合同，自 2025 年 10 月以来锁定至少 14.8 GW 算力。
模型与评测AIHOT feed.xml
aihot.virxact.com打开原文
#42

Research acceleration: The view inside OpenAI

Research acceleration: The view inside OpenAI Apparently today is RSI day at OpenAI, for Recursive Self-Improvement - I think it's their new AGI. Both this piece and the new essay An Alien Mind (by Chief Scientist Jakub Pachocki) talk about it, and this one doesn't even bother to

推荐理由Research acceleration: The view inside：Research acceleration: The view inside OpenAI Apparently today is RSI day at OpenAI。
产品接入Simon Willison
simonwillison.net打开原文
#49

GPT-6 Astra beat Portal start to finish without human help in under 24 hours

GPT-6 Astra beat the puzzle game Portal entirely on its own in about 24 hours, with zero human help after the initial goal was set. Developer cozyblaze published the code and docs on GitHub. His takeaway: Astra is "the worst model we'll ever get." The article GPT-6 Astra beat Por

推荐理由GPT-6：GPT-6 Astra beat the puzzle game Portal entirely on its own in about 24 hours。
Agent与工具行业观察The Decoder
the-decoder.com打开原文
#51

Using Blender with coding agents on macOS

TIL: Using Blender with coding agents on macOS I've been having fun with Blender in ChatGPT Codex on my Mac recently. Getting it to work with coding agents is really easy: install the full Mac application from blender.org and run a prompt like this: Use the already install /Appli

推荐理由Using Blender with coding agents：TIL: Using Blender with coding agents on macOS I've been having fun with Blender。
编码代理Simon Willison
simonwillison.net打开原文