ChatGPT's top pick is more a coin toss than a ranking.问 ChatGPT「哪个最好」，它的首选更像一次抽签，不是一份排名。
In a study published on October 27, the marketing firm BusySeed asked the same question twice and compared 172,624 pairs of answers across 172 categories. With the model version, the web-search setting and the wording all held constant, the brand it placed first matched only 25.4% of the time — three times in four, a different brand came out on top. Asking a week apart or a month apart barely changed that.
The mechanism is plain: a large model writes by sampling tokens probabilistically, so each time you ask, a different name can float to the front. That order reflects how often a brand appears in the training text plus sampling noise, not any judgment about which fits you. There is no stable ranking underneath being read out.
Second-order: as companies start paying to be the brand named first, they are chasing a target that is three-quarters noise; and anyone who treats the top answer as authoritative has handed the comparison step to a process that cannot compare.
The strongest counter: maybe only the wording moves and the set of brands named is stable. That holds for generating a shortlist, but this study measured the first pick, and people act on the first pick.
What it means for you: when you ask an AI for "the best X," treat the answer as a shortlist to check, not a verdict, and compare it yourself against the two or three criteria you actually care about. This judgment needs revising if a replication with the same design shows the top brand holding steady in a clear majority of paired answers.
营销公司 BusySeed 在 10 月 27 日公布的研究里，把同一个问题问两遍、比对了 172,624 对回答，覆盖 172 个品类：模型版本、联网设置、措辞都不变，可排在第一位的品牌只有 25.4% 两次一致——也就是说，四次里有三次，被它推到最前面的是另一个品牌。隔一周问和隔一个月问，差别几乎没有。
机制很直白：大模型是按概率逐字采样生成文字的，同一个问题每跑一次，浮到最前面的那个名字就可能不一样。这个顺序反映的是语料里谁出现得多、加上采样的随机，而不是对「谁更适合你」做过评估。换句话说，底下根本没有一份稳定的排名在被读出来。
二阶效应：当商家开始花钱去抢「被第一个念到」的位置，他们追的其实是一个七成是噪声的目标；而把首选当权威的人，等于把「比较」这一步，交给了一个做不了比较的程序。
最强的反方：也许变的只是措辞，被点到的那一批品牌是稳定的。这对「让它列个候选清单」成立，但这项研究量的正是第一名——而人们是照着第一名行动的。
对你意味着什么：问 AI「最好的 X 是什么」，把它的回答当成一张要去核对的候选清单，不是裁决；拿你真正在意的两三个标准，自己再比一遍。什么会推翻这个判断：如果有人用同样的设计复跑，第一名在明显多数的成对回答里都保持不变，这条判断就要修正。
Built on the WAKENEX concept:所用的 WAKENEX 概念： Match the feedback to the question反馈分工：不要用一种指标替代其他问题的答案
Sources来源
Add WAKENEX as a preferred source on Google在 Google 把 WAKENEX 设为首选来源
READ NEXT接着读
Google's Multi-Party Compute buys stronger key security at one new dependence.谷歌的 Multi-Party Compute 用更强的密钥安全，换来一项新的依赖。
Your coins now move only when Google's cloud agrees to sign. That is the trade worth seeing.
你的币，从此只有在谷歌的云同意签名时才动得了。这就是值得看清的那笔交易。
Open note阅读短评
Most-used on OpenRouter is a price signal, not proof of capability.在 OpenRouter 上用量第一，是价格信号，不是能力的证明。
As of October 27, DeepSeek routes about 46% of the tokens developers send through OpenRouter, more than OpenAI (~12%) and Google (~5%) combined, and the platform's top two models are both the cheap DeepSeek V4.1 Flash variant. The easy read is that DeepSeek is winning the model race. The ranking says something narrower.
截至 10 月 27 日，开发者通过 OpenRouter 发送的 token 里，约 46% 流向 DeepSeek，超过 OpenAI（约 12%）和 Google（约 5%）两家之和；平台排名前两位的，都是便宜的 DeepSeek V4.1 Flash 版本。顺手的读法是『DeepSeek 正在赢下模型竞赛』。但这个排名说的是更窄的一件事。
Open note阅读短评
The Fed's 10-2 hold isn't a verdict that inflation is beaten.美联储 10 比 2 的按兵不动，并不是通胀已被制服的判决。
It's a divided committee buying time. The Committee kept the funds rate at 3.75%-4% while its own statement admits inflation "remains elevated," and two of twelve voters, Lisa Cook and Beth Hammack, pushed to hike another quarter point instead.
它只是一个意见分裂的委员会在拖延时间。委员会把联邦基金利率维持在 3.75% 到 4% 的区间，而它自己的声明却承认通胀『仍然偏高』；十二位票委中有两位——丽莎·库克（Lisa Cook）和贝丝·哈马克（Beth Hammack）——主张应当再加息四分之一个百分点。
Open note阅读短评