#243：GPT-6、Sol 和 Luna、Opus 5.5、新版微软 Copilot、黄仁勋与 AI 末日论者的对决，以及推出 AI Score

但他们是以文明和尊重的方式进行的，这就像……

But they did it in a civil and respectful way, and this is like...

我觉得，这正是AI辩论中所缺失的——就是这种非常开放、诚实的对话，让持不同观点的人能坐在同一个房间里，真正地讨论一些事情。

This is what's missing, I think, in the AI debate is like these very just open, honest dialogues with people with opposing views who can sit in a room and actually discuss something.

说实话，这也是政治中所缺失的。

Like, this is what's missing in politics, honestly.

所以，再次提醒一下，这可是全球最有价值公司的CEO，而他的公司正处于整个AI生态系统的核心位置。

So, again, keep in mind, this is the CEO of the most valuable company in the world, and his company is at the epicenter of the entire AI ecosystem.

说实话，也是当今美国经济的核心。

And honestly, like, the epicenter of today's US economy.

如果没有他们正在做的事情，现在的经济就什么都不是了。

Like, if it wasn't doing what they're doing, the economy is nothing right now.

所以，这……

So that...

这是一件事。

That's one thing.

另外一件事，迈克，就是我上周见到你时跟你提过的那个。

The other, and this is the thing I alluded to you, Mike, I think when I saw you last week.

就是……

The...

这次采访对我来说最重要的收获之一就是，有时候你得亲自去下功夫。

One of the most important takeaways to me with this interview is sometimes you have to do the work as a human.

如果你只看这份文字记录，或者让你的AI帮你总结一下，那你几乎会错过这次采访里所有重要的东西。

If you read the transcript to this only, or if you had your AI do a summary of this transcript, you miss almost everything that was important about this interview.

所以，光靠这个，你没法获得同样深度的理解和背景，你必须亲自去做功课。

So you cannot get the same level of understanding into context without doing the work when it comes to this.

你得亲耳听到黄仁勋说话时的语气。

Like, you have to hear the tone in Jensen's voice.

你得感受到那些挑战，比如那些停顿。

You have to hear the challenges, like the the pauses.

那才是关键所在。

Like, that's everything.

就像你真正理解Jensen的感受，这不是纸上的文字或你的AI助手给你的要点能体现的。

Like the way you actually understand how Jensen feels cannot be seen in words on a page or bullet points that your AI assistant gave you.

你必须亲眼看到他们在想什么、感受什么。

You have to actually see what they're thinking and feeling.

所以不仅仅是听，还要看，这又是另一个层次，因为现在你开始看到背后的情感。

And so not only hearing it, but like watching it is an even another level of it because now you start to see the emotion behind it.

所以给大家一个提醒，要亲自到场。

So just a note for people, like, show up.

要知道，这就是我说的，如果有人派他们的AI代理来参加……

Like, know, this is what I said, like, if someone sends their AI agent to an...

比如和我开会，那我就不跟这个人合作了。

Like a meeting with me ever, like, I'm done working with the person.

如果你不能花时间亲自到场，那就干脆别来。

Like, if if you can't give the time to someone to actually show up and be there yourself, then just don't show up.

别派你的代理来应付。

Like, don't send your agent to do it.

那么……

So...

好。

Okay.

现在我就不再讲大道理了。

Now I'll get off of my, like, preaching.

这是我的一些笔记。

So here is my my my kinda notes.

然后我……

And what I...

我这样做的方式是，我大概花了四天时间在车里听这个采访，然后我开了一个苹果备忘录，我会暂停采访，然后用语音输入记下我的笔记。

The way I did this is I I listened to this interview over probably four days in my car, and then I just had an Apple note going and I would like pause the interview and I would just like voice dictate my notes.

我就直接把这些笔记的原始内容过一遍。

And I'm just gonna kinda go through these notes in their raw.

我甚至都没编辑过这些笔记。

I I haven't even edited these.

所以我认为，可以用一句话来概括詹森的观点：他是个永远的乐观主义者，坚信人类的雄心壮志和辛勤努力会创造一个富足的未来。

So I said, I think you can sum up Jensen's position in that he is an eternal optimist who believes human ambition and hard work will create an abundant future.

其他所有的争论都只是分散注意力，阻碍我们去实现他所描绘的那个未来。

All of the other arguments are just a distraction from creating the future he envisions.

所以，每当伊兹拉遇到挑战时，如果你能回到那个核心点——也就是詹森唯一的关注点，也就是他的信念：一切都会好起来的。

So, like, every challenge Ezra has, if you come back to understanding that that is the sole focus for Jensen, that is his belief is it will work out.

我们只需要保持乐观，并且去亲手创造那个未来。

We just have to be optimistic about it, and we have to create that future.

所以其他那些杂念在他脑海里都烟消云散了。对。

And so all these other things wash away in his mind Yeah.

只要你一直专注于此就行。

If you just stay focused on that.

但当你深入探究那个未来是如何成型的细节时，它本质上就是一种坚信它会实现的信念。

So but when you dig into the details of how that future comes into being, it's literally just a belief that it'll happen.

所以，如果他是个为未来奋斗了三十多年的人，他就相信自己能再次做到。

So, if he is someone who's creating the future for three decades, he believes he can do it again.

所以他就说，我已经做到过了。

So he's like, I've done this.

没人觉得我能把英伟达做成今天这样，但我做到了。

Like nobody thought I could make Nvidia what it is, but I did it.

所以我觉得我还能再来一次。

And so like I can do it again.

所以我喜欢自己做的笔记，就像这样，我在某种程度上和他是一致的。

So I like the note I made to myself, it was just like in this way, I'm kind of aligned with him.

我这些年一直在说，拥抱AI、以AI为先的人，最有可能不仅活下来，还能活得更好。

Like I've been saying for years, the people who embrace AI and become AI forward are gonna have the greatest chance to not only survive, but thrive through it.

我只是更现实一点，那些不拥抱AI的人——我预计未来几年大多数劳动力都会是这样——他们会非常挣扎。

And I'm just more realistic that the people who don't, which I expect to probably be the majority of the workforce over the next few years, are going to struggle tremendously.

所以像Jensen这样的人常常忽略的，往往就是这一部分。

So what people like Jensen just push aside is often that part of it.

所以我对AI以及它可能给社会带来的巨大影响非常乐观。

So I am very optimistic about AI and the impact it could have on society in an incredible way.

科学发现，你知道的，创业方面的进步，给人们自由去创造，给人们更多时间。

Scientific discovery, you know, advancements in entrepreneurship, giving people freedom to, like, build, giving people more time.

这些都是很棒的东西。

Like, there is this amazing stuff.

我也尽量平衡一下，不是对每个人都好，过程中会有很多人被落下，部分原因是他们选择不参与。

I also happen to just try and balance that with it is not gonna be great for everybody, and there's gonna be a whole bunch of people left behind in this process in part because they choose not to participate in it.

而他基本上不提这部分。

And he just basically doesn't talk about that part.

所以，这尤其让我担心现在上大学的人，他们被教导或者自己认为AI纯粹是作弊和抄袭，他们就是拒绝学习它。

So, and that especially becomes a concern for me with, like, the people who are in college now who are taught or either taught or just believe that AI is purely cheating and plagiarism and they just refuse to learn it.

就像，那……

Like, that...

那会是个问题。

That's gonna be a problem.

在考虑AI对工作的影响时，想想工作的目的是什么。

When thinking about AI's impact on jobs, consider what the purpose of the job is.

我觉得这个观点挺有意思的。

I thought this was an interesting perspective.

黄仁勋当时就说，对。

Jensen was like, yeah.

任务会被自动化，但工作的目的不会消失。

Tasks are gonna be automated, but the purpose doesn't leave the job.

所以，他其实也稍微提到了这一点。

So, you know, he did address that a little bit.

然后他谈到了杰文斯悖论，这个所有AI和科技圈的人……

And then he talked about Jevan's paradox, which all the AI and tech...

科技领袖们都喜欢拿出来说，就是AI，你知道，会……

Technology leaders love to throw out there, which is AI just, you know, gets...

随着东西变得更便宜，我们会消费更多。

As things become cheaper, we we consume more of it.

所以一切都会……

So everything's just gonna...

就是说，我们会继续消费更多，然后会有更多的工作被创造出来，诸如此类。

Like, we're just gonna keep consuming more and more jobs will be created, whatever.

当Ezra就新工作的净创造问题向Jensen提出质疑时，你能看出来，这其实并不是一个经过深思熟虑的立场。

When Ezra challenged Jensen on the net creation of new jobs, you could tell that it's not actually a very deeply thought out position at all.

就像他……

Like he...

而这也是我一直跟Sam Altman和其他人争论的。

And this is what I've always argued with like Sam Altman and others.

他们谈论这件事的时候，好像它就会自然而然地发生。

They talk about it as though it's just going to happen.

但当你追问他们，具体怎么发生呢？

And when you push them on, well, how?

他们就说，就是会，因为以前一直是这样。

It's just it will because it always has.

所以他真的……

So he really...

而且，这又是一个听录音特别有帮助的例子，因为他在这点上真的卡壳了。

And again, this is one of those ones where listening actually helps a lot because he really stumbled over this one.

他完全没有什么好的反驳点。

Like he did not have good counterpoints at all.

就是单纯相信人类的雄心能解决一切。

It's just the belief that human ambition will solve things.

当他说AI会创造新东西时，Ezra反驳说，对，但AI不也会擅长做那些新工作吗？

When he said like, well, AI is gonna create these new things, Ezra's point was, well, yeah, but won't AI just be good at doing those jobs too?

我们现在已经到了AI什么都能做得很好的地步，所以就算你创造新工作，理论上AI也会同样擅长那些新工作。

Like, we've now arrived at a realm where the AI is just good at everything And so you create new work, but in theory, the AI is gonna be just as good as the new work you create.

所以我觉得这点挺有意思的。

So I thought that was interesting.

Ezra追问入门级工作的问题，Jensen就插了进来。

Ezra pushed him on the entry level jobs and Jensen jumped in.

有好几次Jensen明显不耐烦了，就直接插话打断Ezra。

There was times where Jensen was obviously getting annoyed, and he would just, like, jump in and talk over Ezra.

他的观点就是，等两年再看。

And his point that was like, just wait two years.

就是说，我们会有一整代新工人，他们是AI原住民，一切都会顺利解决的。

Like, we're gonna have a whole new generation of workers who are AI native and it's just gonna all work out.

我觉得这部分詹森的视角就是不对的。

This is a part where I think Jensen's perspective is just off.

我看这个问题的角度是，他考虑的是工程和AI研究领域。

The way I looked at this one is he's thinking about the world of engineering and AI researchers.

而且，所有进入工程和研究领域的人都会是AI原住民，他们会喜欢这些东西，会直接投入进去。

And, like, everybody who comes up in engineering and research is going to be AI native, and they're gonna love this stuff, and they're gonna just jump in.

但他忘了，未来工人中有很大一部分人讨厌AI，根本不想跟它沾边。

He's forgetting that, like, a large portion of the population of future workers hate AI, like, want nothing to do with it.

所以这非常不受欢迎。

So it's wildly unpopular.

所以你不能假设他们都会成为AI原住民，都会喜欢它，都会觉得这很棒。

And so you can't just assume that they're all just gonna be AI natives and love it, and they're just gonna, like, it's just gonna be amazing.

当他……

When he...

我觉得他对开放模型的观点挺有意思的，因为他是开放权重、开放模型和开源的大力支持者，而我一直觉得这种立场很危险。

I thought it was an interesting perspective on open models because he's a huge proponent of open weights, open models, open source, and I've always thought that's a very dangerous position.

他在这方面并不是一个公正的评论者，我觉得我们必须记住这一点。

He's not an unbiased commentator when it comes to this, and I think we have to keep that in mind.

所以他在年初说，70%的token……

So he said at the start of the year, 70% of tokens...

所以……

So...

而且，再说一次，他的芯片和数据中心正是这些token的动力来源，这些token来自像Claude、ChatGPT、Gemini等封闭模型。

And, again, his chips, his data centers are what power these tokens, were coming from closed models like Claude and ChatGPT, Gemini, etcetera.

但现在情况反过来了，70%的计算资源现在用于开放权重模型。

But that that is now flipped and 70% of the compute is now going to open weight models.

不管怎样，他们都能赚钱。

They make money either way.

所以他不在乎是开放模型还是封闭模型。

So he doesn't care if they're open models or closed models.

他们反正都要为英伟达的芯片付钱。

They're still paying for NVIDIA chips.

所以他……

So he...

他就像，让我们……

He's like, let's...

这都挺好的。

This is all good.

关于Hugging Face事件，这倒是个有趣的视角，因为现在他拥有Hugging Face。

On the Hugging Face incident, which is an interesting perspective because he now owns Hugging Face.

记得吗，他们两周前以129亿美元收购了它。

Remember, they bought them two weeks ago for $12,900,000,000.

是的。

Yep.

顺便说一句，我以前从来没听说过，Hugging Face的CEO克莱姆去找黄仁勋说，嘿。

By the way, I did not hear the story ever before that Clem, the CEO of Hugging Face, went to Jensen and said, hey.

把我们买了吧。

Buy us.

而且他在采访里多少透露了这件事。

And he kinda disclosed that on the interview.

但他淡化了当时的情况，这就是你说的，迈克。

But he downplayed what was going on, and this is where you said, Mike.

他基本上就是说，这全都是软件。

He basically just says, this is all just software.

这全都是工程问题。

It is all an engineering problem.

这些智能体集群，没什么新鲜的。

These agent swarms, they're nothing new.

就像我们早就有了并行计算什么的。

Like we've had parallel computing and everything.

埃兹拉在这个问题上反驳了很多次，但他们总是绕回这个话题。

Ezra pushed back on this one a lot and they kept coming back to this.

但詹森基本上觉得，听着，这是个工程问题。

But Jensen, you know, basically felt like, listen, it's an engineering problem.

他们搞砸了。

They screwed up.

他们推出了一个糟糕的产品。

They put a bad product out.

埃兹拉说，他们根本没推出产品。

Ezra was like, they didn't put a product out.

关键在于这发生在测试环境中。

That's the point is this happened in a testing environment.

然后詹森说，那他们就是搞砸了，他们会修好的。

And he said, well, then they screwed up and they'll fix it.

所以詹森非常坚持认为，这个问题一定会被解决，因为推出安全产品的动力非常大，否则他们就要为这些事承担责任，如果他们不这么做的话。

So Jensen was very much like, it's just gonna get solved because there's tremendous motivation to put out safe products because otherwise they're gonna be liable for these things, that if if they don't.

然后他说的那句话，大家都拿来当标题，就是‘如果他们控制不了，那就把实验室关了’。

And then he did, you know, the sound bite everybody ran with is like, well, if they can't contain them, then they shut their labs down.

结果这就成了大家……

Like, that became the thing everybody...

但再说一次，你得听完整段对话才能明白。

But, again, you needed the full context of the conversation.

你得听出Jensen语气里的那种不耐烦，他不是真的说要关掉Anthropic和OpenAI。

You needed to hear the, annoyance in Jensen's voice that he's not literally saying shut down Anthropic and OpenAI.

这就像……

It's like a...

这几乎是一种修辞，就是说‘如果他们搞不定，那就干脆关掉算了’。

It's almost like a figure speech of like, well, if they can't, then just shut it down.

他知道那并不是真实会发生的情况。

He knows that that's not an actual scenario.

他并不是在提议谁去关掉任何实验室。

He's not proposing anybody shuts any labs down.

他谈了很多关于监管的事情，以及为什么他认为现有的监管已经足够了。

He talked a lot about regulation and why he thinks that the existing regulation is sufficient.

然后他回应了Ezra不断打断的情况，但就像，这件事发生了，实验室以为他们知道发生了什么。

He then addressed like Ezra kept breaking up, but like, but this happened and the labs thought that they knew what was going on.

这让我想起了上周Noam Brown的评论，Mike，因为我实际上还在……

Now it brought me back to the Noam Brown comments from last week, Mike, because I'm still actually...

Noam在接受Dwarkash采访时承认，他们在Hugging Face事件中没有进行思维链监控。

Noam admitted in the interview he did with Dwarkash that they did not have chain of thought monitoring on the hugging face incidents.

对。

Yep.

我不是律师，但这看起来像是在公开承认疏忽。

I am not an attorney, but that seemed to be admitting negligence, very publicly.

我敢打赌，这已经出现在即将提交的针对OpenAI模型造成损害的诉讼的法律文件中。

And I got to imagine that is already in a legal brief that is going to be filed in a lawsuit against OpenAI for damages done by their models.

他们确实犯了大错。

They did make major mistakes.

就像，他们……

Like, they...

有些事情，即使作为一个……

There are things that even as an...

我有……

I have...

这不是我的领域。

This is not my realm.

我不是AI研究员。

I am not an AI researcher.

我不训练这些模型。

I don't train these models.

当他说那话时，我心想，什么？

When he said that, was like, what?

什么意思？

Like, what do you mean?

你根本没有监控思维链。

You weren't monitoring chain of thought.

那可是用来对齐模型的关键。

Like, it is the thing that allows you to align the models.

所以对我来说，我有点同意詹森的看法。

So that that to me is like, I'm kinda with Jensen.

比如，什么……

Like, what...

OpenAI 发生的那些乱七八糟的事情，看起来本来是可以控制的，结果他们搞砸了。

All this crazy stuff that happened with OpenAI seems like it was very manageable, and they screwed up.

而且他们一直承认自己搞砸了，总有一天这会反过来在法律上咬他们一口。

And they have continually admitted it to screwing up, and it has to come back and bite them legally at some point.

他坚信那只是软件问题，这点我不同意，但他很坚持这个观点。

He believes it's just software, which I don't agree with, but he's big on that.

另一个我觉得值得一提的点是，他们提到了中国，因为很明显，詹森在特朗普政府那里得到了青睐，以至于他们能把芯片重新卖回中国，而很多人反对把我们最强大的芯片卖给中国。

The other one I thought that was worth note is they brought up China because, obviously, Jensen's been, gotten favor with the Trump administration to the point where they could get their chips back into China, and that's a lot of people oppose the idea of putting our most powerful chips into China.

你得记住黄仁勋是怎么看待竞争的。

You have to remember how Jensen thinks about competition.

所以，迈克，为了做到这一点，我其实，你知道，我想给谷歌的Gemini和它在Google Drive里的集成点个赞。

And so to do this, Mike, I actually, you know, I guess I'll give a shout out to to Google on Gemini and the integration in Google Drive.

我就在Google Drive里搜了一下，找到了那期节目，因为我们所有的节目文字稿都在Google Drive里，那期我们聊了黄仁勋关于竞争的理论。

I just searched in Google Drive, find the episode, because all of our show transcripts are in Google Drive, where we talked about Jensen's theories on competition.

然后它找到了我在2024年10月第119期节目里做的笔记，那期我们引用了黄仁勋在BG2播客采访中的话。

And it found my notes from episode one nineteen of October 2024, where we quoted Jensen from his BG, two podcast interview.

这就是他对竞争的总体理念。

So this is his overall philosophy on competition.

他当时说，作为一家公司，我们要保持对局势的……

He said at that time, as a company, we want to be situation...

对局势的敏锐感知，我对我们公司以及我们生态系统的方方面面都非常敏锐。

Situationally aware, and I'm very situationally aware of everything around our company and our ecosystem.

我了解所有在做不同事情的人，以及他们在做什么。

I'm aware of all the people doing alternative things and what they're doing.

有时候对我们有敌意，有时候没有，但我非常清楚。

And sometimes it's adversarial to us, sometimes it's not, I'm super aware of it.

但这不会改变公司的使命。

But that doesn't change the purpose of the company.

我们不是要从谁那里抢市场份额。

We're not trying to take share from anybody.

英伟达是市场创造者，不是市场份额的争夺者。

NVIDIA is a market maker, not a share taker.

看看我们公司的幻灯片，我们从来没有一天谈论过市场份额，内部也从不谈这个。

If you look at our company slides, we don't show not one day does this company talk about market share, not on the inside.

我们谈论的只是如何创造下一个新事物？

All we're talking about is how do we create the next thing?

下一个问题是什么？

What's the next problem?

换句话说，他根本不在乎竞争。

In other words, he doesn't care about the competition.

他觉得他是个终极乐观主义者。

His feeling is he's an ultimate optimist.

他是个工程师，他相信自己会一直领先所有人。

He is an engineer, and he believes he will stay ahead of everybody.

所以他的想法是，把芯片给中国。

So his feeling is give China the chips.

谁在乎呢？

Who cares?

我们只会比他们做得更好。

We will just be better than them.

我们会创新得更快。

We will innovate faster.

所以我觉得，总的来说，这就是黄仁勋的态度——我们总能找到解决办法。

And so I think, like, big picture that sums up Jensen's take that, like, we will just figure this out.

我们能解决一切问题。

We will solve everything.

如果我们做对了，采取正确的策略，我们就能解决这个问题。

And if we do it and we take the right approach, we'll figure out.

再说回黄仁勋，今天早上他们做了什么？

Now to Jensen's credit again, what do they do this morning?

他们发布了公告，成立了OpenAgent安全平台，专门解决这个问题。

They dropped the announcement that they have formed the OpenAgent safety platform to solve this exact problem.

他在推特上发了消息，而且黄仁勋上X平台才两个月左右。

So he tweeted, and again, Jensen's only been on on X for like two months.

他今天早上发布说，今天我们与超过100家行业伙伴一起，推出了NVIDIA OpenAgent安全平台，整合了OpenShell和Sentry这两个内部工具。

He put out this morning, today with over a 100 industry partners, we introduced NVIDIA OpenAgent Safety Platform, bringing together OpenShell and Sentry, two internal things.

人工智能是一项非凡的技术，它将推动未来几代人在发现、生产力、安全、健康和繁荣方面取得进步，但只有当人们相信AI是以安全的方式构建，并以智慧和责任来部署时，它的全部潜力才能实现。

Artificial intelligence is extraordinary, technology that will advance discovery, productivity, security, health, and prosperity for generations to come, but its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.

这不仅仅是一个产品。

This is bigger than a single product.

这是一个开放生态系统的开端，旨在为安全的智能体系统建立信任层。

It's the beginning of an open ecosystem to build the trust layer for safe agent systems.

我们一起在打造AI经济的基础。

Together, we are building the foundation of the AI economy.

信任和创新并不冲突。

Trust and innovation are not in conflict.

安全是赢得信任的方式。

Safety is how trust is earned.

我们不仅要打造最强大的AI，还要打造最值得信赖的AI，这样这项非凡的技术才能为世界实现其巨大的潜力。

We must build not only the most capable AI, but the most trusted so this extraordinary technology can realize its enormous promise for the world.

所以他在采访中说这是个工程问题，然后他通过推出一个生态系统或平台来支持自己的说法，试图解决采访中强调的那个问题。

So he does an interview, says it's an engineering problem, then he backs it up by putting out an ecosystem or a platform to try and solve for the exact problem that was highlighted on the interview.

所以，本着尽可能乐观和解决问题的精神，虽然我在采访中对黄仁勋的很多观点并不认同。

So again, in the in the spirit of focusing on as much we can optimism and solving problems, I think why I don't agree with Jensen on quite a number of things from the interview.

但我尊重他把它看作一个可解决的问题，并且他会尽一切努力去解决它。

I respect that he sees it as a solvable problem, and he's gonna do everything in their power to try and solve it.

我喜欢这一点。

I love that.

好的。

Alright.

这是我们这周的第三个大话题。

Our third big topic this week.

保罗，上周你介绍了一本我们讨论过的新书，《AI转型蓝图》。

So, Paul, last week, you introduced a new book that we discussed, the AI transformation blueprint.

然后这周……

And this fast...

这周你分享了一个叫SmarterX的AI评分工具的首次亮相。

This past week, you shared a first look at a tool called AI Score by SmarterX.

你是在LinkedIn上分享的。

So you had shared this on LinkedIn.

这个工具整合了两项评估：面向组织的商业AI评分和面向个人的个人AI评分。

This tool brings together two assessments, business AI score for organizations and personal AI score for individuals.

所以我想请你多给我们讲讲SmarterX的AI评分工具。

So I'd love for you to just tell us a little bit more about AI Score by SmarterX.

这个项目已经筹备了相当长一段时间了。

This has been in the works for quite a while.

你能给我们详细介绍一下吗？

Can you walk us through it?

首先，先说一下它的可用性。

So first, just like a heads up on availability.

这个我确实在LinkedIn上宣布过，也在上个周末的ExecAI通讯里提到了。

So this is something I did announce on LinkedIn, and I put it in my newsletter, my ExecAI newsletter this past weekend.

我们实际上会在MACON上首次亮相。

We are actually gonna be debuting this at MACON.

从10月13日到15日，它会正式开始推出。

So October thirteenth to fifteenth is when it'll kind of officially start rolling out.

然后我会举办一个网络研讨会。

And then I'm gonna hold a webinar.

我记得是10月28日。

I think it's October 28.

我们会在节目说明里放上链接，介绍整个产品，然后正式推出，并且说明它是免费的，这样大家就能用了。

We'll put the the link in the show notes, sort of explaining the whole product and, like, officially rolling it out and showing it's gonna be free, so people can use it.

所以目前大概就是这样。

So just kind of that's where it's at.

这会是一个公开测试版。

It'll be a public beta.

我今天其实在做最后的内部测试。

I'm actually doing final beta internal testing today.

就是我们希望推出的最终版本。

The the kind of the final version that we're hoping to roll out.

然后我们会……

And then we'll...

我们可能会在未来一周左右，把它扩展成一个更长时间的私人测试。

We might put it into like a extended, private beta over the next week or so.

其实我在过去七八个月的播客里，可能已经提到过AI Score大概有五六次了。

So I have actually alluded to AI Score, probably a half dozen times over the last seven or eight months on, this podcast.

我之前提到的很多应用场景，其实都和开发 AI Score 有关。

A lot of the use cases I've talked about were related to the building of AI Score.

简单来说，事情是这样的：2006 年初，我开始负责一个内部项目，叫作“AI 转型系统”。

So, basically, what happened is in early two thousand six, I started working on an internal initiative called the AI transformation system.

那个……

What...

它的背景是，我们当时有……

Where it comes from is, we have...

我们的 AI Academy 已经有成百上千家企业客户，我也花了不少时间和这些组织里的很多领导者交流，还和其他行业的高管们做过私下沟通。

So our AI Academy has hundreds of business accounts, and I have spent time with a lot of leaders at those organizations, had private discussions with executives across a bunch of other industries.

经过对比发现，一个非常明显的情况是，企业和领导层都在为如何定义转型的样子而发愁，比如怎么设定基准线、明确现状，然后找到一种方法来衡量转型随时间的进展。

And what's become compare, very apparent is, like, organizations and leaders are struggling to define what transformation looks like, benchmark, set baselines where they are today, and then have a way to measure progress of that transformation over time.

所以，AI Score 就成了这套转型体系的基础。

So AI Score became the foundation of that transformation system.

我觉得它更像是一个变革管理工具，能让领导层对组织进行评估。

So I think of it as a change management tool that lets leaders assess organizations.

所以有两个版本。

So there's two versions.

一个是企业AI评分，用来评估整个组织。

That's Business AI Score, where you assess the organization.

我觉得这个特别适合那些推动转型的领导者，他们对组织在这些不同领域的实力最有洞察。

I see that being ideally suited for the leaders who are driving the transformation, who have, like, the most insight into the strength of the organization across these different areas.

但同时也很有帮助的是，获取其他利益相关者的看法，比如业务负责人、团队主管等等，因为他们的视角可能和全球AI转型负责人非常不同。

But also very helpful to get other stakeholder perspectives like business leaders and team leads and things like that, because their perspective may be very different than the global head of AI transformation.

然后个人方面，iScore让个人自我评估。

And then personally, iScore lets individuals assess themselves.

所以整个理念是，企业AI转型需要在这些不同的支柱和维度上采取特定的方法。

So the whole idea is business AI transformation requires a certain approach across these different pillars and dimensions.

而每个企业AI转型基本上就是一系列个人转型的集合。

And then every business AI transformation is basically a collection of individual transformations.

所以个人就在这些不同的支柱上自我评估。

And so individuals then assess themselves across these different pillars.

所以，一旦你拿到AI评分结果，就会用这些结果来指导制定个性化的学习路径和转型蓝图。

And so what happens is once you get your AI Score results, you then use those to inform the building of personalized learning journeys and transformation blueprints.

AI评分是基于这样一个理念：转型需要深度的变革管理。

So AI Score is based on the idea that transformation requires deep change management.

我们看到的，再说一次，我和数百家组织交流过。

What we have seen, again, I've talked with hundreds of organizations.

过去三年，大多数组织进行转型的方式是，他们去弄来Copilot、Claude、ChatGPT或Gemini，然后交给员工，而员工缺乏使用这些工具的上下文和理解。

The way most organizations for the last three years have approached transformation is they go and get Copilot Claude, Chateappity or Gemini, then they give it to their employees who lack context and understanding of how to use the tools.

所以基本上，转型就是这样进行的。

And so that's basically what transformation has been.

他们给员工工具，然后希望他们自己摸索出使用方法。

They give them the tools and then hope they figure out how to use them.

所以我在2月26日开始构建AI评分评估，先从商业AI评分开始，然后，你知道，用同样的流程构建了个人AI评分。

So I started work in February 26 on building AI Score assessments And I started with Business AI Score, then I, you know, use the same process to build Personal AI Score.

所以从背景来看，这是我之前实际执行过的一套方案。

So contextually, this is a playbook I've actually run before.

所以，在我拥有自己公司的时候，如果你是新听众，我经营了一家咨询公司，一家营销咨询公司，长达十六年。

So back when I owned my agency, and if you're new to the show, like, I owned a consulting firm, a marketing consulting firm for sixteen years.

那是HubSpot在07年的第一个合作伙伴。

That was HubSpot's first partner back in 'seven.

所以，我当年就建了一个评估工具，我记得大概是2011年左右，它成了我们市场策略的基础，也是我们所有客户合作的依据。

So I built an assessment tool back in the day, I think it was like 2011 or so, that became the foundation of our go to market strategy and the basis for all our client engagements.

那个工具评估营销人才、技术和策略，然后你看优势和劣势，再围绕这些劣势制定计划。

Now that tool assessed marketing talent, tech, and strategy, and then you looked at the strengths and weaknesses, and then you built your plans around those weaknesses.

所以，在我心里，我们一直计划做这个。

So in the back of my mind, we always plan to do this.

其实，我几年前建过一个叫“营销人员AI评分”的工具，后来停用了。

Now I actually built one that we sunset years ago called AI Score for Marketers.

所以，如果你读过我们那本关于营销和人工智能的书……

So if you read our marketing, artificial intelligence book...

那本书对吗？

Was that was that the right book?

那是2022年的书。

That was the book in 2022.

对。

Yep.

我们确实写了关于AI Score for Marketers的内容。

We actually wrote about AI Score for Marketers.

所以我几年前就停用了那个，然后一直打算把AI Score重新建造成一个更通用的工具，适用于各种企业。

So I sunset that a couple years back and then always intended to, like, rebuild AI Score as a more general purpose tool for for businesses.

但当初我建那些工具的时候，还没有AI能帮我做原型。

But what I didn't have back when I built those other ones was AI that could help me build prototypes of these things.

我不是开发者。

So I'm not a developer.

我没办法自己做出真正的原型。

I can't build the actual prototype.

所以我花了很多时间，自己构建评估体系，定义支柱、维度这些内容。

So I spent all this time, building the assessments themselves, defining the pillars, the dimensions, all this stuff.

但现在我可以去Claude那边说，好，帮我做出这些工具的功能原型。

But now I could go into Claude and be like, okay, help me build functioning prototypes of these things.

所以我会在这上面花更多时间，比如我们做网络研讨会和正式发布的时候，但简单来说，每个工具都有这些支柱。

So I'll spend a lot more time on this, like when we do the webinar and the official launch, but high level, each of them has these pillars.

商业AI的支柱，我们其实聊过这些。

So the business AI pillars, we actually talked about these.

我只是没把它们放到AI Score工具的框架里。

I just didn't put them in the context of the AI Score tool.

这些支柱是愿景、战略、数据、技术、治理、素养、人员和绩效。

So those are vision, strategy, data, technology, governance, literacy, people, and performance.

这就是商业AI Score的八大支柱。

Those are the eight pillars for the business AI Score.

然后个人AI Score是知识，比如你懂不懂AI？

And then the personal AI Score is knowledge, like do you understand AI?

应用，你能不能把它用起来？

Application, are you able to put it to work?

判断力，你会负责任地使用它吗？

Judgment, do you use it responsibly?

而且当AI输出结果时，你基本上知道什么算是好的吗？

And do you know what good looks like basically when AI outputs something?

影响力，它是否让你在工作中表现更好？

Impact, so is it making you better at your job?

然后是心态，你是否准备好持续成长？

And then mindset, so are you built to keep growing?

所以AI Score的理念是，你完成这些评估，得到你的分数，不仅整体上，而且在各个支柱上，都会给你一个弱、中、强的等级。

And so the idea with AI Score is you take these, you get your score, gives you kind of a weak, moderate, strong grade across not only overall, but then the pillars.

然后很酷的一点是，企业管理员可以邀请团队成员也来参加评估。

And then the cool thing is business admins will be able to interview or, invite team members to also take it.

然后他们会得到一个管理仪表板，这样他们就能看到我如何评价组织与Mike如何创建组织之间的差异。

And then they'll get an admin dashboard so they can actually then look at the difference between like how I graded the organization and how Mike created the organization.

所以它旨在真正成为一个变革管理工具。

And so that it's meant to truly be a change management tool.

再说一遍，这些全都是免费使用的。

Again, that's all gonna be free to use.

所以，简单来说，就是AI的使用，这周我的AI应用案例就是这个，Mike，你可以以后慢慢做。

So, like, just high level, like, use of AI, and this will be my AI use case for the week, Mike, so you can just do years later on doing.

所以我们……

So we...

一旦我建好原型……

Once I build the prototypes...

所以总共，我从二月到现在，大概花了三四百个小时，做评估，创建所有评估，然后从那里，用Claude把这些变成HTML原型。

So in total, I probably spent about three to four hundred hours from February until, I mean, up to today, doing the assessments, creating all of the assessments, and then from there, turning those into HTML prototypes using Claude.

我主要用的是Sonnet 4.6。

Now I use mostly, like, Sonnet 4.6.

然后当Fable 5.1出来的时候，我开始用Fable 5和5.1，我开始付费买额外的积分。

And then once Fable 5.1 came out, I started, and Fable five and five point one, I started paying for extra credits.

然后我开始迭代原型，我建了几十个这样的原型。

And I started building iterations of the prototypes, which I built like dozens of these prototypes.

然后我把它交给了实际的软件开发团队，他们正在全力把这个产品推向市场。

And then I turned it over to actual software development team who is, sprinting this thing to market.

我们大概会在三十天内从我的原型做到最终的生产版本。

Like, we're gonna go from my prototypes to final production in about thirty days.

哇。

Wow.

再说一次，因为以前做过类似的事情，我自己用 vibe coding 做原型，可能为 SmarterX 省了大约十万到二十五万美元。

Again, having done this before, I probably saved SmarterX somewhere between a 100,000 and $250,000 by vibe coding the prototypes myself.

等我发布完整的可用套件时，你会明白这些工作的深度。

And when I release, like, the full suite of what's available, you'll realize, like, the depth to all of this work.

但我估计，我们不仅可能省了二十五万美元，而且时间上也会轻松多花一倍，而且我绝不可能在今年秋天 MACON 之前及时推出这些。

But I would estimate not only we do we save probably a $250,000, it would have taken twice as long easily, and I would have never launched these in time for MACON in, like, this fall.

所以，是的，我的意思是，从高层次来看，这是我做过的最紧张的事情，就花费的时间而言，但它也让我看到了未来工作和组织架构的样子，这有助于我为五月的主题演讲提供信息，就像，是不是……

So, yeah, I mean, at a really high level, it it it is, it's the most intense thing I've I've ever done in terms of the time it took, but it also opened my eyes to the future of work and org charts, which helps inform the keynote I'm doing at May kind of like, is it...

我们的团队结构会是什么样子？

What our team structures look like?

因为实际情况是，我在设计整个系统，但同时也在自己动手搭建，这在以前我是做不到的。

Because the reality is like, I'm architecting this entire thing, but I'm also building the whole thing myself, which I wouldn't have had the the ability to do.

这开始改变你对工作如何压缩的看法，以及以前是领导者的人现在会成为建设者，他们能够做很多事情。

And it starts to change the way you think about how work is gonna compress and how people that were leaders before are now gonna be builders and they're gonna be able to do a lot of the things.

我甚至做了很多关于工具发布的市场推广工作，因为我只需要跟我的助手说一声，就能开始把这些东西做出来。

And I've even done a lot of the go to market work around the launch of the tool because I just talked to my assistant and I can start building that stuff out.

所以你可以去 smarterx.ai/score，那里会带你加入等待名单。

So you can go to, smarterx.ai/score and that'll take you to join the wait list for it.

再说一次，它可能会在10月13日发布，我觉得大概就是这个日期。

Again, it'll it'll probably launch October 13, I I think is probably the date.

如果我觉得准备好了，可能会提前一点发布，但最有可能的是13号出来。

I I might drop it a little bit early if I think it's ready, but most likely the thirteenth is when it'll come out.

然后我们会举办那个网络研讨会，暂定名为“面向商业领袖的AI评分”，时间是10月28日，届时我们会全面介绍这个平台以及如何使用它。

And then we'll do that webinar, AI Score for Business Leaders is the tentative name for the webinar on October 28, where we'll sort of give the whole overview of the platform, how to use it.

所以我真的很兴奋。

So I'm really excited.

说实话，在幕后默默做几个月的事情，却完全不能提，真的挺难受的，尤其是当你真的相信它会带来很大的影响，比如那种以人为本的AI应用方式。

It's like, it's hard to work on something sort of behind the scenes for months and not talk about it at all, especially when, you know, you really believe it's gonna make a big impact on, like, a human centered approach to this stuff.

不过，终于能聊这些了，感觉还挺好的。

But, yeah, it's cool to finally be able to talk about this stuff.

我觉得，这本书加上这个工具，真的几乎就是一个绝佳的公开构建案例，展示了现在AI能实现什么。

I mean, between this and the book, it's really almost a incredible case study of building in public and what's possible now with AI.

我特别喜欢这一点。

I love that facet of it.

而且书里确实更深入地讲述了这整个过程的来龙去脉。

And the book does tell the story a little bit deeper about the whole process behind this.

所以，其实书的第一部分就是讲这一切是怎么开始的。

So I actually like, you know, the first section of the book is sort of how it all came to be.

第二部分是商业AI转型，以及AI Score如何推动这个转型。

And then section two is the business AI transformation and how AI Score powers that.

然后，书的第三部分是个人AI转型，以及如何在这个过程中使用个人AI Score。

And then, the the third section of the book is the personal AI transformation and how to use personal AI Score within that.

所以，是的，就像我说的，我们把这些都免费开放了。

So, yeah, and like I said, we're just making it all free.

这些东西，如果我只是想开个咨询公司，我们就会把这些都保密，然后靠这个做几百万美元的项目。

Like, this is the stuff that if I just wanted to build a consulting firm, we would have kept all this, like, proprietary, and we would have just done million dollar engagements based on this stuff.

但这不是我们的商业模式。

But that's not our business model.

我们的商业模式是加速人们对这些东西的理解和负责任的应用。

Our business model is let's accelerate understanding and responsible adoption of this stuff.

所以我们的做法就是把工具和知识分享出去，然后尽快帮助尽可能多的人。

So our approach is let's just put the tools and knowledge out into the world, and then, you know, try and help as many people as we can as quick as we can.

好的。

Alright.

在我们进入快速问答环节之前，保罗，这期节目还要感谢Outshift，也就是思科的前沿技术孵化引擎。

Before we dive into our rapid fire, Paul, this episode is also brought to you by Outshift, which is Cisco's incubation engine for Frontier Technology.

现在，多智能体AI已经出现在每家企业的路线图上了。

So multi agent AI is in every enterprise roadmap right now.

但这是每个人都在面对的问题。

But here's the problem that everyone is facing.

智能体可以传递消息，但不能一起思考。

Agents can pass messages, but they can't think together.

系统会因认知失误而默默表现不佳，比如对已验证声明的误读和虚假共识，这些是传统监控无法发现的。

The system silently underperform through cognitive failures like misreadings on verified claims and false consensus that conventional monitoring cannot detect.

现在，思科的Outshift正在构建解决方案，他们称之为“认知互联网”。

Now Outshift by Cisco is building the fix, what they call the Internet of Cognition.

这是一个开源基础，用于构建从设计到生产的多智能体系统，具备共享上下文、共享记忆和防护措施，以实现我们对AI的真正期望。

This is an open source foundation for building multi agent systems from design to production with shared context, shared memory, and guardrails to drive the results we actually expect from AI.

你可以通过阅读论文、体验演示和从outshift.com获取代码来了解更多。

You can go learn more about this by reading the paper, experiencing the demo, and grabbing the code from outshift.com.

那就是outshift.com。

That is outshift.com.

好的，保罗。

Alright, Paul.

我们进入快问快答环节吧。

Let's dive into rapid fire.

最近我们一直在关注的一个大话题是，OpenAI计划淘汰自定义GPT，并引导大家转向插件。

Big topic that's been on our minds lately is that OpenAI is planning to retire custom GPTs and move people towards plug ins.

他们目前的FAQ里说，这个转变会影响所有ChatGPT套餐。

Its current FAQ on this says the transition affects all chat GPT plans.

不过，迁移的权限和用户可用的选项可能会因账户或工作区而有所不同。

Although migration access and the options available to users can differ by account or workspace.

根据我们目前掌握的信息，自定义GPT似乎计划在2026年12月11日退役。

Custom GPTs, far as based on the information we have right now, appear to be scheduled to retire on 12/11/2026.

对于受影响的工作区，OpenAI现在计划在10月26日停止创建新的GPT。

For affected workspaces, OpenAI now plans to stop new GPT creation on October 26.

现有的GPT在各自的退役日期之前仍然可以使用，OpenAI表示用户应该留意他们在账户或工作区中收到的通知。

Existing GPTs remain usable until their applicable retirement date, and OpenAI says users should follow the notices they receive in their account or workspace.

所以他们是在推动大家把GPT变成插件，迁移到插件，并且以后都用插件。

So they are pushing people towards turning your GPTs into plug ins, migrating them to plug ins, and using plug ins moving forward.

插件打包了可重复使用的指令和连接工具，供ChatGPT和编解码器使用。

Plug ins package reusable instructions and connected tools for chat GPT and codecs.

所以在这种计划中的迁移里，GPT的指令会变成一个技能。

So in this kind of planned migration, a GPT's instructions become a skill.

它的知识文件会变成插件的参考文件，而连接的应用程序基本上会移入插件中。

Its knowledge files become reference files for the plug in, and its connected apps basically move into the plugin.

GPT选定的模型、现有的对话和自定义操作不会转移。

The GPT selected model, existing conversations, and custom actions do not transfer.

任何依赖自定义操作的工作流都需要一个替代集成，才能再次正常工作。

Any workflow that depends on a custom action needs a replacement integration before that part will work again.

迁移还会使用GPT的最新发布版本，所以草稿和未发布的更改会被排除在外。

Migration also uses the latest published version of the GPT, so drafts and unpublished changes are left out.

他们有一个流程，你可以开始迁移现有的GPT。

They have a process by which you can start migrating your existing GPTs.

有趣的是，我们会讨论这个。

Interestingly, we'll talk about this.

当你把GPT迁移到插件或者创建新插件时，插件默认是私有的。

The plug in starts as private when you migrate your GPTs to a plug in or when you create a new one.

原有的共享设置不会保留。

Existing sharing settings do not carry over.

所以你们组织里的用户需要访问这个替代插件。

So users in your organization will need access to the replacement.

看起来目前还没有简单的方法来公开分享插件。

It does not appear that there's an easy way to share plugins publicly.

我们稍后会再详细聊聊这个。

We'll talk a little bit more about that.

迁移后，原来的GPT会变成只读状态。

The original GPT becomes read only after migration.

OpenAI明确提醒说，插件可能会有不同的响应，所以创建者需要测试它们。

OpenAI fully warns that plugins may respond differently, so creators need to test them.

那么，保罗，你想从哪里开始呢？

So, Paul, where do you wanna start here?

这看起来是个相当大的变化。

This seems like a fairly big change.

我不知道OpenAI是否认为这个变化对某些听众以及依赖GPTs的非技术用户来说有多大。

I don't know if OpenAI sees this as as big of a change as it might be for some of our audience and more nontechnical people who rely on GPTs.

我的意思是，通常来说，他们并不在意。

I mean, usually, it's the sense they don't care.

比如，不。

Like, no.

那真是太疯狂了。

That's pretty wild.

他们确实不在意。

They do not.

所以，这里说一下背景。

So context here.

OpenAI在开发者日上推出了GPTs。

So OpenAI introduced GPTs during dev day.

所以，再说一遍，今年9月29号是开发者日。

So, again, September 29 is the dev day this year.

他们是在2023年11月的开发者日上推出的，就在山姆·奥特曼被解雇前11天，后来他又被重新聘为CEO。

So they introduced them Dev Day November 2023, eleven days before Sam Altman was fired before being rehired as the CEO.

我猜这个时间点，这种紧迫感，跟今年的开发者日有关。

I'm guessing the timing, like the urgency of this, has something to do with this year's dev day.

所以他们，对，

So they're Yeah.

可能会宣布一大堆东西，他们需要先把这些推出来。

Probably announcing a bunch of stuff, and they needed to just get this out there.

奇怪的是，GPT让创建和分享自定义AI助手变得极其容易，用ChatGPT做AI助手也变得超级简单。

The the part that's weird is GPT's made AI assistants and working with ChatGPT extremely easy, to create and share these custom AI assistants.

这本来是个很神奇的东西，但他们似乎从来没有真正兑现最初对自定义GPT的愿景，比如建立整个市场之类的。

It was like a supernatural thing, but they just never seemed to commit to the original vision for custom GPTs that they shared, like building this entire marketplace and things like that.

所以我一直不明白他们为什么支持不够，这让我总觉得，总有一天他们会把这些东西砍掉，因为他们根本没有像预期那样去支持它们。

And so I I just never understood their lack of support, which always led me to believe, like, at some point, they're just gonna get rid of these things because they're just not supporting them the way you would expect.

我教过几十个基于GPT的工作坊和课程，比如我们的工作GPT、问题GPT、创新GPT，这些东西对人们来说很有意义。

Now having taught dozens of workshops and courses based on GPTs that I built, so our jobs GPT, problems GPT, innovations GPT, these things made sense to people.

它们让AI对非技术用户来说更易接近、更实用。

They made AI more approachable and actionable for those non technical users.

你走进一个房间，打开工作GPT，展示一下，大家立刻就明白了。

You get in a room, you bring up jobs GPT, you show it and the light bulb goes off.

所以这就是为什么我……

And so that's why I just...

我一直不明白他们为什么不更多地推广这些，因为在我看来，这是教别人在不知道如何利用AI时最快获得价值的方法。

I never understood why they didn't lean into them more because to me, it was like the fastest way to teach someone how to get value out of AI when they didn't know what to do with it.

所以有点疯狂。

So kinda wild.

所以12月11日看起来是所有自定义GPT的预定退役日期。

So December 11 looks like the scheduled retirement date for all custom GPTs.

它们将在那天停止工作。

They'll just stop working on that date.

我在X上发过一条帖子，说等OpenAI解决了所有千禧年大奖难题之后，我希望他们能推出一个未发布的模型，来解释一下插件、应用、技能和GPTs这些品牌概念的区别，而GPTs最终会变成插件。

I, posted on X, when OpenAI is done solving all the millennium prize problems, I ask that they turn their unreleased model at explaining the branding function around plugins versus apps versus skills versus GPTs, which will become plugins.

他们本来有GPTs，这东西容易理解、容易构建、容易分享，结果他们把它给搞砸了。

They took GPTs, which were easy to understand, build, and share and blew it up.

所以，如果Mike之前给你解释的那些还不够让你困惑的话，那我就直接给你读一下OpenAI发布的ChatGPT插件页面开头的内容。

So if what Mike explained to you already wasn't confusing enough, I'm just going to read to you the opening to the plugins in ChatGPT page that OpenAI published.

这是原文。

This is verbatim.

插件通过打包可复用的指令、连接的工具和其他功能，帮助ChatGPT和Codex完成工作流程。

Plugins help ChatGPT and Codex complete workflows by packaging reusable instructions, connected tools, and other capabilities.

一个插件可以包含提供指令和工作流指导的技能、连接外部账户、信息和操作的应用，以及帮助管理员为工作区配置应用的模板。

A plugin can include skills that provide instructions and workflow guidance, connected apps that link external accounts, information and actions, app templates that help administrators configure apps for a workspace.

有些插件包含多个应用。

Some plugins include multiple apps.

其他插件只包含技能，不需要连接应用。

Others include only skills and do not require an app connection.

应用和插件的作用是不同的。

Apps and plugins serve different purposes.

现在，如果你真的去尝试做这件事，会变得超级混乱，因为如果你今天去OpenAI里构建一个代理，也许他们会修复这个问题，你可以连接……

Now this gets super confusing if you actually try and do this because if you go in to build an agent in OpenAI today, maybe they fix this, you can connect...

我可能自己都搞不清楚。

I might not even get this right.

你可以把一个应用连接到代理上，对。

You can connect an app to an agent Yeah.

但是你不能把一个插件连接到代理上。

But you can't connect a plug in to an agent.

所以，比如，如果我把Jobs GPT重新做成插件，我在构建代理的时候就没法调用它了。

So, like, I can't call jobs GPT if I rebuild it as a plug in when I'm building an agent.

这简直让人抓狂，就像我和Mike，当这个公告出来的时候，我们真的坐在那里讨论了大概四十五分钟。

It's insanely confusing to like Mike and I, when this announcement, we literally sat there for like forty five minutes.

这到底是怎么回事？

Like what the hell?

这跟他们宣传的方式完全对不上。

Like this doesn't even make any sense for how they're promoting this.

好了，回到正题。

So anyway, back to it.

所以，应用和插件用途不同。

So apps and plugins serve different purposes.

你看到的选项取决于你的账户、工作区以及你使用它们的地方。

The options you see depend on your account, workspace and where you use them.

所以这甚至不是一套通用的术语。

So it's not even like a universal terminology.

现有的应用连接仍然受它们自己的授权和工作区控制的约束。

Existing app connections remain subject to their own authorization and workspace controls.

安装插件可能会启动应用连接或设置流程。

Installing a plugin may start an app connection or setup flow.

你仍然必须完成任何必要的账户授权，安装不能绕过提供商或工作区的权限。

You must still complete any required account authorization and installation cannot bypass provider or workspace permissions.

然后你想起这个，天哪，我超喜欢这个。

And then you recall this one, oh my god, which I love.

迁移后的插件在广泛分享之前可能会有不同的表现，先比较一下熟悉的提示词，再试一个更难的情况。

A migrated plugin may respond differently before sharing it more widely, compare familiar prompts and at least one harder case.

检查它是否选择了正确的技能，并遵循你的指示。

Check that it selects the right skill and follows your instructions.

这一点非常重要。

That's pretty important.

使用预期的参考资料。

Uses the expected reference material.

这样它就不会拉入你不想让插件引用的其他知识文件，而显然你是在给其他人访问权限的，它能生成完整的答案或文件，具备工作流所需的工具和集成，解决影响工作流的差异。

So it doesn't like pull some other knowledge file that you don't want pulled into the plugin that apparently you're giving access to other people, produces complete answers or files, has the tools and integrations the workflow needs, resolves differences that affect the workflow.

然后他们有一些问答，我能和同样的人继续分享我的GPT吗？

Then they have some Q and A, can I share, keep sharing my GPT with the same people?

基本上不行，对，我就简短说，不行。

No, is the basically, yeah, I'll just shorten that one, no.

我能用别人的GPT吗？

Can I use someone else's GPT?

基本上答案就是不行。

No is basically the answer to that one.

那这让你处于什么境地呢？

So and where does that leave you?

就是说，我们该怎么处理这一切？

Like, what do we do with all this?

因为我们有一堆GPT，不仅是我们公司运营的核心，我们还教这些东西。

Because we have a bunch of GPTs that are core to not only running our company, but we teach these things.

所以基本上我觉得有三个选择。

So basically I see it as like three options.

你可以直接用他们的API，把你的GPT重新做成独立的应用程序或者技术。

You can just, you know, use their API and just rebuild your GPTs as like standalone apps or, you know, technology.

你可以创建一个插件并分享，如果插件允许分享的话，不过我们对此不是百分百确定。

You can create and share a plugin maybe if sharing's allowed with plugins, which we're not a 100% clear on.

或者你直接把提示词给别人，说，拿去用吧，想怎么弄就怎么弄，这大概就是我们现在的方向，其实我们一直都是这么做的。

Or you just give people the prompts, here, do whatever you want with this, which is kinda like what we're leaning towards, which was sort of what we've always done.

就是说，把提示词公开，让别人自己去搭建，因为我们教工作坊和课程的时候，经常用ChatGPT的GPTs。

It's like, just be open about what the prompt is and let somebody go build it because we teach our workshops and our courses often using like chat GPT- GPTs.

但我们发现，有人可能会在Copilot、Claude或者Gemini里用这些。

But we realize like people might take it that are in Copilot or Claude or Gemini.

所以我们经常就直接说，这是提示词。

So we'll often just be like, here's the prompt.

拿去，想怎么用就怎么用。

Like, just go do whatever you want.

就自己搭一个这样的助手。

Like, just build this kind of assistant.

所以我和Mike其实打算做……

So Mike and I are actually gonna do...

我们准备在我们的AI学院里推出一个新东西，叫AI Show Extended。

We're we're kinda launching a new, thing in our AI Academy called AI Show Extended.

所以我们接下来要做的是，把播客里那些没时间深入讲的核心概念，拿到AI学院里做成课程来详细讲。

So what we're gonna start doing is taking some of these core concepts from the podcast that we don't have time to go into deeply, and we're just gonna do, courses in AI Academy about them.

所以我和迈克，虽然还不确定最终名字，但我们想出来的名字是“GPT的WTF”，就是这门课的名字。

So Mike and I, I don't know if we're gonna go with this final name, but the name that Mike and I came up with was GPT- GPT's WTF was like the name, it's the course.

团队可能会给它改个名。

It'll probably get renamed by the team.

但我们会录一门课，更深入地讲这个。

But we're going to record a course that goes deeper into this.

然后短期内，在我们的课程里也会涉及。

And then we're also in the near term in our courses.

所以如果你是学院的会员或客户，你会看到我们会在点播课程里加一个视频，来解释这个。

So if you're an academy, member or customer, you'll see we're gonna actually install a video that goes into the on demand courses that sort of explains this.

然后如果课程里没有提供提示词，我们就会直接把提示词给大家，比如“嘿，自己重建一个吧”。

And then we're gonna, if prompts aren't ready, aren't offered in the courses, we're just gonna give people the prompts like, hey, rebuild your own into...

在我们重新打造课程的同时。

While we're kind of recreating our courses.

然后我们会用最新的内容来更新课程。

And then we'll update our courses with the current thing.

但关键是这些东西会不断变化。

But the key is they're gonna keep changing.

比如说，人们……

Like, people...

他们会……

They get...

人们可能会生气，也许三个月后他们又会重新推出GPTs。

People are gonna get pissed, they might reintroduce GPTs in three months.

谁知道呢？

Like, who knows?

所以我们一直围绕框架来构建课程，而不是试图针对特定技术。

And so we've always built our courses around frameworks, not like trying to be technology specific.

所以我们早就知道这种情况可能发生，他们可以像昨天早上或周五早上对Copilot那样改变用户体验。

And so we've known that this stuff can happen and they can change the user experiences like they did with Copilot just, you know, yesterday morning or Friday morning.

所以我们尽量让内容保持常青、可扩展，但我们也正在转向更动态的学习模式，更注重实时内容和即时内容，尤其是在平台本身方面。

So we try and make our stuff as evergreen and scalable as possible, but we're also moving to a more dynamic learning model where we're we're focused more on, like, live content and real time content, especially when it comes to the platforms themselves.

保罗，这里还有几点要补充，可能只是重申你刚才说的。

And a few other notes here, Paul, that probably just reiterate what you had said.

所以，那些比普通知识工作者更懂行的人可能会说，当然，你可以通过创建插件或技能来获得同样的功能。

So, you know, people out there who are much more savvy than your average knowledge worker might be saying something like, well, of course, you can get the same functionality by creating a plugin or a skill.

这说得有道理，也可能是真的，但这不是重点。

And it's like, that's fair and possibly true, but it's not the point.

重点是，现在我必须经历所有这些额外的步骤，这些步骤我认为更加……

The point is now I have to go through all these other steps that are much more, I would argue, much...

更加繁琐。

Have more friction.

我得去创建一个技能。

I have to go create a skill.

我得把技能和文件打包在一起。

I have to package the skill up with files.

我得搞清楚怎么调用应用，然后把这一切打包成一个插件。

I have to figure out how to call apps, I suppose, and then package this all into a plug in.

你可以通过他们的插件创建器用自然语言来完成。

You can do this via natural language via their plug in creator.

这还是有工作量的。

It's still work.

这比创建自定义GPT要麻烦得多，那个只需要两秒钟。

It's much more onerous than creating a custom GPT was, which took two seconds.

你可以很快搞定。

You can do very quickly.

我建议所有听众，如果你想尝试，第一步就是……

I would encourage everyone listening as your first step if you wanna experiment.

说实话，我会先创建技能。

I would frankly just create skills first of things.

其他的那些东西，先别管。

Like, don't even bother with all this other stuff yet.

如果你还没用上技能功能，其实就是一些指向文件夹、示例和文件的Markdown文件，那就从那里开始，然后再把它变成插件。

Just like if you're not using skills yet, which are just markdown files that point at folders and examples and files, start there and then turn that into a plug in.

暂时别担心应用那部分，除非真的非用不可，那样的话你就得自己搞定了。

Don't even worry about the app piece of it yet unless it's absolutely critical, then you're gonna have to figure that out.

不过我知道很多人已经创建了不少GPT。

But there's plenty of GPTs I know people have created.

我们内部也创建了一些。

We've created internally.

它们基本上就是，给你一些指令。

They're basically just like, here's instructions.

去参考一些文件，希望你能很快上手，比如可以测试的东西。

Go reference some files that hopefully you can get to pretty quickly, like something you can test.

现在还不确定它是否还能像以前那样工作，但至少对大家来说是个起点。

Now jury's out on whether it'll still work the same way, but that's at least a start for folks.

但接着就会遇到一个问题，比如你可以和团队分享这个。

But then it runs into this whole thing of, like, you can share this with your team.

你可以和你的团队分享一个插件。

You can share a plug in with your team.

很好。

Great.

这是一种完全不同的用户体验和功能形式，所以这会造成摩擦。

It's a totally different user experience and form function, so that's friction.

你不能公开分享这些。

You cannot share these publicly.

你可以提交你的插件给OpenAI审核，让它公开。

You can submit your plugin for an OpenAI review process to make it public.

最终，他们不会给出任何时间表。

Eventually, they don't give any timelines for this.

你通过插件提交报告提交它……

You submit it through a plug in submission report...

门户。

Portal.

OpenAI会审核提交的内容……

OpenAI reviews the submit...

说实话，这个功能的存在本身就完全违背了创建公开GPT的初衷。

Like, the fact this exists functionally just defeats the entire purpose of creating public GPT.

所以这一点非常重要，需要理解。

So that's a huge piece to understand.

我知道不是每个人都在做公开的GPT，但如果你一直在依赖这些，我在开发者论坛上看到过一些评论，有人说，我是为我的客户创建这些的。

I realize not everyone does public GPTs, but if you have been relying on these, I've seen comments on the developer boards and things of people being like, I create these for my clients.

那解决方案是什么呢？

Like, what is the solution?

答案是，你没办法了，对，就是这样。

And the solution is you're screwed Yeah.

至少目前是这样。

Right now, I would say.

抱歉。

Sorry.

但确实是这样。

But, like, that's true.

我的意思是，我……

I mean, I...

我只是随便看了一眼。

Like, I'm just glancing.

我做的那个求职GPT已经有四万四千次对话了，真的被广泛使用。

So my jobs GPT when I built has been 44,000 conversations, like, it's been used in.

超过八百条评价。

Over 800 ratings.

所以，这个工具对大家真的帮助很大。

So, I mean, that's a tool that's just been super helpful for people.

我知道大家经常用它，而他们现在就要把它关掉了。

I know people use it all the time, and they're just shutting it down.

所以它……

So it...