Mollick 只给 Opus 说了一句'你看着办'，模型自己把活全干了，这条演示能帮你重新理解提示词怎么写。
宾夕法尼亚大学研究者 Ethan Mollick 演示只用一句提示词——要求 Opus 把内容做成电影并“自己想办法”——就让 Claude Opus 自主完成后续多步骤生成。Opus 自行拆解任务、串联工具链，无需用户逐步指导。这展示了一句话委托式提示词与逐步指令式工作流的差别。
"Opus, great, now turn that into a movie, i know you can make it amazing. figure it out."
Yes, that was the only prompt https://t.co/ZmlVl4pcse
Simon 又发观察文了：他说 Opus 5.5 连 3D、音乐、游戏设计都能一手包办，GPT-6 Astra 暂时还差点趣味感，做游戏的可以看看。
Google 两周内连发两个 Gemini 4，新出的 Carbon 专攻找漏洞、打补丁，综合分 59 排到 Artificial Analysis 榜首，安全运维方向可以关注。
作者用 Claude Opus 搭了个短视频生产流水线，从解析素材到出三套方案再到自动生成，全程 Agent 跑完，想搭内容工作流的可以看看。
一个搞安全的推友观察到：找漏洞 DeepSeek 和 Opus 打平，但修老系统还得靠 Opus，聊聊算力怎么分配。
The 54-point gap between GPT-5 and GPT-6 Astra in about a year shows how quickly frontier models are absorbing research-level physics when given code execution. Tracking these scores before the late October Intelligence Index update helps researchers and benchmark watchers pick models for scientific workflows. Compare against Claude Opus 5.5 at 78% when choosing a frontier model for agentic computation tasks.
The scaffolding, not the model, roughly doubled the improvement — a design lesson for anyone building AI-scientist or agent systems. Builders of research agents can copy the three skill modules (ideation, experiment management, scheduled checklists) and benchmark against Agent4Sci's 20 tasks. The paper also exposes predictable agent failure modes, such as wasted compute and narrow local sweeps, that teams can check for directly.
