OpenAI 一週內抄出競品 API：根本沒訓練新模型

核心論點 · 點時間戳可跳到原聲

原話 · 已逐字校驗

they said that a few months ago, I think, and I hope they have a different perspective now because Computer Use is, like, 180 degrees different than it was.

他們幾個月前是這麼說的，我希望他們現在有不一樣的看法了，因為 Computer Use 已經和當時完全不一樣了，可以說是 180 度大轉變。

Ari Weinstein4:53

now Computer Use, often writes code. So if you actually look at it in Codex and you expand the tool calls manually, you can see that it’s not just doing one action at a time. It’s actually writing JavaScript code that it executes

現在 Computer Use 經常是直接寫代碼。如果你在 Codex 裡把工具調用展開看，會發現它不是一步一步執行單個動作，而是真的在寫一段 JavaScript 代碼去執行。

Ari Weinstein5:57

if you take a screenshot of a webpage that has a link- The screenshot doesn’t include where the link goes. It doesn’t include, you know, maybe you take a screenshot of your calendar, the ca- event ti- titles are truncated, you know?

如果你截一張帶鏈接的網頁截圖，截圖裡是看不到鏈接指向哪裡的。也看不到完整信息——比如截一張日曆的圖，事件標題是被截斷的，你懂吧？

Ari Weinstein11:27

now Computer Use is, like, faster at accomplishing tasks than, like, the average human probably in most cases. and I think that the next frontier is to have Computer Use be, like, literally superhuman in its performance

現在 Computer Use 完成任務的速度，在大多數情況下已經比普通人更快了。我認為下一個前沿是讓 Computer Use 真正變得超人類——性能上比我們這樣的電腦操作專家還要快。

Ari Weinstein12:32

One of my favorite use cases for Computer Use actually, and one that we see a lot in the wild, is Computer Use letting the agent- actually test the software that the agent has built, which is far more consequential than it sounds.

我最喜歡的 Computer Use 用例之一，也是我們在實際使用中看到很多的，就是讓 Computer Use 去測試 agent 自己構建出來的軟件，這件事的意義比聽上去重要得多。

Ari Weinstein17:31

This is, like, really just Luna. And, on top of that, what you’re doing is you’re constraining. So, like, structured output’s a big part of it. you’re really optimizing the inference stack to, like, get very fast on TTFD.

這其實就是 Luna。在這基礎上，你做的事是加約束——結構化輸出是很重要的一部分，你在推理棧上做極致優化，讓首 token 時間（TTFD）變得非常快。

Nikunj Handa27:26

We’ve been, trying to, like, really up our game on caching. We provide now guarantees of, like, cache hits within, like, 30 minutes. We’re actually, like, we-- for one of our users, we just launched, like, a much longer cache window. So we have, like, a 12-hour caching guarantee

我們一直在努力把緩存這件事做得更好。現在我們提供 30 分鐘內緩存命中的保證。實際上我們剛剛為一位客戶上線了更長的緩存窗口——我們現在有 12 小時的緩存保證。

Nikunj Handa32:12

數字與實體

術語

App Shot應用快照

雙擊快捷鍵抓取應用完整無障礙數據，而非普通截圖

TTFT/TTFD首Token延遲

模型生成第一個輸出token所需時間，衡量推理速度的關鍵指標

mid-turn steering推理中途引導

在模型推理過程中途插入新指令來調整其行為

WebSocketsWebSocket雙向連接

支持異步工具調用和實時交互的持久通信協議

收聽指南