6 Questions Every Enterprise Has to Answer About AI — 文稿

M10:00

今天的 AI Daily Brief，我们来聊六个正在塑造企业 AI 的问题。

Today on the AI Daily Brief, 6 questions shaping enterprise AI.

在那之前，先看头条：Sam Altman 去了 Washington，而过去这一周，相关讨论已经变得复杂多了。

Before that, in the headlines, Sam Altman goes to Washington, and the conversation has gotten a lot more complicated over the last week.

M10:51

Sam Altman 现在已经到 Washington，去见议员和白宫官员了。而上周一开始敲定这趟行程的时候，议程其实还挺简单的。

Well, Sam Altman has arrived in Washington to meet with lawmakers and White House officials, and when the trip was set at the beginning of last week, the agenda was pretty simple.

Altman 本来是要向 Washington 介绍 OpenAI 新模型的能力，并讨论一套发布流程，希望避免再出现 o3 和 GPT-5.6 发布时那样的情况。

Altman would brief Washington on the capabilities of OpenAI's new model and discuss a protocol for release, hopefully avoiding a repeat of the o3 and GPT-5.6 rollout.

但从那之后，OpenAI Hugging Face hack 发生了，围绕 OpenWeights 模型也出现了公开争论，还有一份很吸引眼球的请愿书，呼吁政府出手，建立能够放慢前沿 AI 发展速度的能力。

Since then, however, we've had the OpenAI Hugging Face hack, a public debate about OpenWeights models, and an attention-getting petition for the government to step in and build the capability to slow down the pace of frontier AI.

换句话说，短短几周之内，Altman 要面对的讨论已经复杂了很多。

In other words, conversations have become a lot more complicated for Altman in just a couple of weeks.

根据报道，Altman 在周三见了 Senate Commerce Chair Ted Cruz 和几位民主党参议员，但外界几乎没得到什么关于具体谈了什么的信息。

According to reports, Altman met with Senate Commerce Chair Ted Cruz and several Democratic senators on Wednesday, but we got very little information on what was actually discussed.

Altman 对记者表示，他不愿说明这个正在被预览的模型会在什么时候发布，甚至是否会发布都不愿确认，只是说了一句，不确定。

Speaking to reporters, Altman declined to state when or even whether the model being previewed would be released, commenting, not sure.

而这正是我们今天要聊的部分。

That's the part we're here to talk about.

Altman 也拒绝谈这个模型有哪些新的能力会引发担忧。

Altman also declined to discuss the new capabilities of the model that give cause for concern.

当然，Hugging Face 那起事件在这次访问中影响很大，不过现在越来越像是，处在那场争议中心的那个模型不会被发布了。

Now, of course, the Hugging Face incident looms large over this visit, but it increasingly appears like the model at the center of that controversy will not see release.

OpenAI 在周二更新他们的 postmortem blog 时表示，那个模型只是一个内部研究原型，本来就从没打算对外发布。

In a Tuesday update to their postmortem blog, OpenAI said that the model was an internal-only research prototype never intended for public release.

在 Washington，Altman 对媒体说，这个模型现在已经被永久停用，连内部研究也无法再访问了。这也就意味着，大概率它并不是这周向议员展示预览的那个模型。

In Washington, Altman told the press that the model has now been permanently deactivated and is inaccessible even for internal research, meaning presumably it's not the model being previewed to lawmakers this week.

Sam 还说，他和 Ted Cruz 并没有讨论具体立法，不过他说，引用一下，我们谈了我们的新模型，也谈了 America 要想在 AI 上保持竞争力，需要做到什么。

Now, Sam said that he and Ted Cruz had not discussed specific legislation, but that, quote, we talked about our new model and what it's going to take for America to remain competitive with AI.

Altman 还表示，他不支持强制性的安全测试，尤其是因为这可能会给 OpenWeights 模型开发者带来不必要的负担。不过他也补充说，对于能力达到新水平的 frontier models，我们认为联邦政府具备很强的测试能力和测试手段，真的非常重要。

Altman also said that he didn't support mandatory safety testing, particularly because it could introduce an unnecessary burden on OpenWeights model developers, but added, for frontier models at new levels of capabilities, we think it's really important that the federal government has great testing capacity and capabilities.

Altman 说，他计划在本周晚些时候再去见一批其他官员，其中包括白宫幕僚长 Susie Wiles。无论出于什么原因，她现在已经成了 AI 政策方面的关键决策者之一。

Altman said he plans to meet with a range of other officials to end the week, including White House Chief of Staff Susie Wiles, who, for whatever reason, has wound up as one of the key decision makers on AI policy.

那场会面很可能会谈到自愿性的 AI 安全测试框架，而这个框架的截止日期是八月一日。

That meeting will likely include a discussion of the voluntary AI safety testing framework, which has a deadline of August 1st.

报道称，这套框架已经发给 OpenAI、Anthropic 和 Google 征求意见了，但 Altman 拒绝对这份草案发表评论。

Reports state that this framework has been circulated to OpenAI, Anthropic, and Google for comment, but Altman declined to comment on the draft.

在 Capitol Hill 的一场走廊采访中，Altman 被问到，他会不会和白宫谈一谈是否有必要放慢 AI 开发速度。

In a hallway interview on Capitol Hill, Altman was asked whether he would talk to the White House about the need to decelerate AI development.

Altman 代表他团队最近那封公开信里的观点回应说，我不会用 deceleration 这个词，但随着模型能力越来越强，我们会讨论需要把握节奏，我觉得这符合所有人的利益。

Representing the views of his staff from the recent open letter, Altman responded, I wouldn't use the word deceleration, but we talk about the need to pace it as the models get more capable, which I think is in everyone's interests.

另外还有一件事，我明天会更深入地讲。

Now, one other story that I'm going to get into in more depth tomorrow.

那就是 OpenAI 和 Anthropic 两边的营收数字都出现了显著增长。

Is the significant increase in revenue numbers on both the OpenAI and Anthropic front.

不过我不想把这条消息埋没在头条里，所以它会成为明天的重点话题。

But I do not want to bury that in the headline, so that will be a major topic for tomorrow.

到时候记得回来听。

Come back for that.

简单来说，OpenAI 的 CFO Sarah Friar 最近告诉员工，七月份的年化营收已经超过了整个上一季度的水平。

Suffice it to say, the CFO of OpenAI, Sarah Friar, recently told employees that annualized revenue in July topped all of the previous quarter.

再补一条和 OpenAI 有关的悬念消息。

One more bit of OpenAI intrigue.

总裁 Greg Brockman 说，公司正在研发一整套设备，好让他们的聊天机器人拥有实体存在。

President Greg Brockman says that the company is working on an entire range of devices to give a physical presence to their chatbots.

在接受前 Wall Street Journal 记者 Joanna Stern 的一场新采访时，Brockman 确认 OpenAI 的硬件计划仍然在按计划推进，并表示公司正在打造一个设备家族。

In a new interview with former Wall Street Journal reporter Joanna Stern, Brockman confirmed that OpenAI's hardware plans are still on track, stating that the company is building a family of devices.

他没有确认最近传闻中的 smart speaker，也没有确认今年外界猜测过的其他设备形态；至于时间表，他也没有多说，只是表示，你们很快就会看到。

He wouldn't confirm the recently rumored smart speaker or any other form factors that have seen speculation this year, nor would he give a timeline beyond commenting, you can expect them soon.

不过，这已经是我们目前听到的最明确的确认了：完整的硬件产品线仍然在路线图上，哪怕经历了 io 的终结以及来自 Apple 的一场知识产权诉讼，它也还在。

Still, this is the clearest confirmation we've had so far that a full hardware range is still on the roadmap, surviving the end of io and an IP lawsuit from Apple.

Brockman 在谈到那场诉讼的时候，很可以理解地说得很简短，只表示：我们专注于自己的研发和技术。

Brockman was understandably brief when talking about that lawsuit, stating, we are focused on our own development and technology.

还有一条挺有意思的竞争动态，我觉得这对消费者来说是个好消息，尤其是对你们这些在企业里、对要用哪些模型和平台其实没太多选择的人来说。Microsoft 看起来正准备通过开发一个 Copilot super app，更直接地和 OpenAI 以及 Anthropic 竞争。

One interesting bit of competitive news, which I think sounds good for consumers, particularly those of you who are in the enterprise without a ton of choice on which models and platforms you're going to use, Microsoft appears to be gearing up to compete more directly with OpenAI and Anthropic with the development of a Copilot super app.

在周三晚上的财报电话会上，CEO Satya Nadella 确认，这个 app 会在今年晚些时候推出，目标是把面向消费者和企业客户的 Copilot 体验统一起来。

During Wednesday night's earnings call, CEO Satya Nadella confirmed the app is coming later this year with the goal of unifying the Copilot experience for both consumer and enterprise customers.

他说，Copilot 正在迅速从 chat 演进到 co-work，再到 autopilots。

He said, Copilot is rapidly evolving from chat to co-work to autopilots.

这个季度，我们会把这些 Copilot 体验整合到一起，包括 code，放进一个 super app 里。

This quarter, we are bringing these Copilot experiences together, including code, in one super app.

这是一个重大的前进一步，我也很期待很快能分享更多信息。

This is a major step forward, and I look forward to sharing more soon.

由于他们新的 MAI models 的能力，Microsoft 开始把 OpenAI 和 Anthropic 视为直接竞争对手。

Microsoft is beginning to see OpenAI and Anthropic as direct rivals, thanks to the capabilities of their new MAI models.

Nadella 对分析师表示，成本和数据隐私方面的担忧结合在一起，给了 Microsoft 一个机会，去说服客户采用他们自己更便宜的模型。

Nadella told analysts that the combination of cost and data privacy concerns gives Microsoft an opportunity to sell customers on their own cheaper models.

当被问到关于 open 和 closed 之争这个持续不断的讨论时，Nadella 觉得这种框架过于简单化了。

When asked about the rolling debate about open versus closed, Nadella suggested the framing is too simplified.

他说，目标是让公司能够掌控自己的命运。

He said, the goal is to have the firm be in control of their own destiny.

我们对这个平台的架构设计是非常、非常明确的，那就是你可以让自己的 harness 和 model 保持分离。

We are very, very clear about the architectural design of the platform, which is you get to keep your harness separate from the model.

这意味着，在任何一个时间点，任何 model 都可以替换。

That means any model at any given time is swappable.

现在，我知道你们当中有些人会觉得这有点像公司式的回答，但我其实觉得，他对大多数企业心态的判断是对的，因为我不认为大多数企业最终真的在乎一个 model 到底是 open 还是 closed。

Now, I'm sure some of you will think that that's kind of a corporate answer, but I actually think that his assessment of how most enterprises feel is correct in that I don't think that most enterprises actually care ultimately about whether a model is open or closed.

他们在乎的是，自己能拿它做什么，自己掌握多少控制权，以及为了获得他们正在使用的这些系统的访问权，他们要在控制权上向别人让渡多少。

They care what they can do with it, what control they have, and what sacrifices around control they're making to someone else to have access to the systems they're using.

总之，整体来看，Microsoft 越来越把自己定位成一个 model-agnostic 的平台，提供全套选择，而不是 OpenAI 或 Anthropic 产品的转售商。

Anyway, overall, Microsoft is increasingly positioning themselves not as a reseller of OpenAI or Anthropic products, but rather as a model-agnostic platform offering a full range of options.

Nadella 说，每个客户都希望针对每一项任务，基于 latency、quality、cost 和 compliance，选到合适的 model。

Said Nadella, every customer wants the right model for each task based on latency, quality, cost, and compliance.

我们在云上提供最广泛的 model catalog，拥有超过一万一千个 models，其中包括 OpenAI、Anthropic、Mistral、xAI 的领先模型，也包括我们自己的 MAI family。

We offer the broadest model catalog in the cloud with over 11,000 models, including the leaders from OpenAI, Anthropic, Mistral, xAI, as well as our own MAI family.

在 AI 领域这些重大讨论和争夺位置的混战之中，Mark Zuckerberg 又在 The Wall Street Journal 的一篇新署名文章里，为加速 AI 发展提出了自己的理由。

Now, along the swirl of all these big discussions and jockeying for position in AI, Mark Zuckerberg has made the case for AI acceleration in a new op-ed in The Wall Street Journal.

在一篇题为《The AI Future Is for Everyone》的文章中，Zuckerberg 认为，AI 时代那个决定性的问题，不会是 superintelligence 会不会存在，而是谁能够接触到它。

In an essay titled The AI Future Is for Everyone, Zuckerberg argued that the defining question of the AI age won't be whether superintelligence will exist, but who will have access to it.

换句话说，就是我们最终会进入一个怎样的世界：是 superintelligence 被少数几个机构紧紧掌握，还是被广泛分发给普通人。

In other words, whether we end up in a world where superintelligence is closely held by a handful of institutions or broadly distributed to normal people.

Zuckerberg 写道，令人惊讶的是，很多正在开发 artificial intelligence 的人，他们的话语里竟然充满了末日论调。

Zuckerberg wrote, it is surprising that the discourse from many of those who are developing artificial intelligence is so filled with doom.

我不明白，为什么那些相信 AI 会消灭大多数工作、也会让人类的大部分意义变得无关紧要的人，会急着去建设那样的未来。

I don't understand why anyone who believes that AI will eliminate most jobs and much of humanity's relevance would rush to build that future.

说句题外话，这其实正是我昨天在讨论我认为普通人会如何回应那封 Pacing the Frontier 信时，想表达的核心观点：对于“你们为什么要造 AI”这个问题，唯一能接受的答案不能是，“嗯，如果我们不做，别人也会做。”

This is, for what it's worth, exactly the point that I was trying to make yesterday when I was discussing what I think the normie response to the Pacing the Frontier letter would be, that the only acceptable answer to why are you building AI is not, well, if we don't, someone else will.

而应该是，因为我们觉得 AI 会非常棒，而且它带来的好处会远远超过随之而来的所有风险。

But instead because we think AI will be awesome and dramatically better than all the risks that it comes with.

Zuckerberg 接着说，那种认为 AI 危险到唯一安全的道路就是极端集权的想法，本身就很危险。

Zuckerberg continued, the notion that AI is so dangerous that the only safe path is an extreme concentration of power seems dangerous.

从历史上看，寄希望于某种绝对权力在足够开明的情况下会仁慈地照顾人类，并没有带来安全或积极的结果。

Historically, hoping that an absolute power will benevolently provide for humanity if sufficiently enlightened hasn't led to safe or positive outcomes.

Zuckerberg 的看法是，就像之前的那些技术，比如 internet，最好的结果会来自把这项技术自由地扩散到整个社会。

Zuckerberg's view is that much like previous technologies, like the internet, the best result will come from diffusing the technology freely across society.

他写道，我们相信，回答这个问题的方法，不是把这种力量集中起来，而是把 personal superintelligence 交到每一个人手中。

He wrote, Rather than centralizing this power, we believe that delivering personal superintelligence to everyone is the way to answer this question.

这有可能开启一个个人赋权的新时代，在那个时代，个体会拥有更大的自由去追求自己的兴趣，并充分发挥自己的潜能。

This has the potential to begin a new era of personal empowerment, in which individuals have greater freedom to pursue their interests and reach their full potential.

这篇署名文章也是一次媒体宣传行程的一部分，并且和 Meta 新推出的 AI 乐观主义宣传活动配套联动。

Now, the op-ed came as part of a press tour that linked up with Meta's new AI optimism campaign.

在接受 The Wall Street Journal 的另一场采访时，Zuckerberg 呼吁美国政府加快 AI 发展，而不是去限制它。

In a separate interview with The Wall Street Journal, Zuckerberg called for the US government to accelerate AI development rather than restrict it.

他的观点是，广泛分发 AI 带来的好处，远远大于它的风险。他还补充说，我知道，关于未来的讨论一直都很难，因为那些事还没发生。但我确实认为，到现在这个阶段，我们已经有很多数据点了，而这些数据点应该让我们比当前舆论所反映出来的态度乐观得多。

He argued that the benefits of broadly distributing AI outweigh the risks by quite a margin, adding, I get that it's always hard to debate about the future because it hasn't happened yet, but I do think we have a lot of data points at this point, and that should point us to be much more optimistic than I believe the current discourse reflects.

具体来说，他警告说，不要觉得政府三十天或六十天的审核窗口是无害的。他评论说，这个领域发展得太快了，所以这实际上是一段相当有分量的时间。

Specifically, he warned against thinking a 30- or 60-day government review window is harmless, commenting, the field is moving so quickly that actually is quite a meaningful amount of time.

值得注意的是，Meta 是唯一一家还没有同意政府自愿测试框架的前沿 AI 实验室。

Now notably, Meta is the only frontier AI lab that hasn't agreed to the government's voluntary testing framework.

Zuckerberg 还在接受 Financial Times 的另一场采访时表示，美国政府不应该封禁中国的 AI。

Zuckerberg also said that the US government shouldn't ban Chinese AI in a separate interview with the Financial Times.

他不只是认为封禁不会有效，他还相信，这会带来 regulatory capture 的风险，而且更广泛地说，可能会阻碍 open models 的发布。

Not only does he think a ban won't be effective, but he believes it would open the risk of regulatory capture and could stymie the release of open models more generally.

另外，Meta 的首席 AI 官 Alexandr Wang 最近表示，公司将再次开始推出 open-source models，这也说明这不只是空喊口号。

Now, Meta's chief AI officer, Alexandr Wang, recently said that the company will begin launching open-source models again, suggesting that this isn't just hollow sentiment.

不过，总的来说，核心信息其实很简单，就是我们需要对 AI 更乐观一些。

Still, overall, the core message is simply that more AI optimism is needed.

在接受 The New York Times 采访时，Zuckerberg 说，很多其他正在开发这项技术的实验室，他们的讨论里都充满了压倒性的末日论调。

Speaking with The New York Times, Zuckerberg said, so much of the discourse from a lot of the other labs that are developing this is overwhelmingly filled with doom.

这场辩论里需要有一个声音，或者几个声音，把现实感带进来。

There needs to be a voice or several voices that are bringing realism to this debate.

不过我觉得，很遗憾的是，Mark Zuckerberg 作为 AI 乐观主义主要代表人物的影响力，会受到历史因素以及人们对社交媒体整体社会影响相当负面的看法所限制。

Now, I think unfortunately Mark Zuckerberg's power to be the leading face of AI optimism is limited by history and people's fairly negative view of the overall impact of social media on society.

但即便如此，开始形成一种高调而且持续的讨论，让其他人能够接过去继续推动，这一点非常重要。你也完全可以相信，我会在这里放大这个信息。

Still, to start to have loud, sustained discourse that other people can pick up and run with is immensely important, and you better believe I will be here amplifying that message.

不过现在，今天的头条内容就先到这里。

For now, however, that's going to do it for today's headlines.

接下来，进入今天的主节目。

Next up, the main episode.

现在关于 AI 最重要的问题之一，不是谁在用 AI，而是谁把它用得好。

One of the most important AI questions right now isn't who's using AI—it's who's using it well.

M110:14

这就是在一个超大型企业代码库里做开发的真相。

Here's the truth about building inside a massive enterprise codebase.

写代码从来都不是瓶颈——上下文才是。

Writing code was never the bottleneck— context is.

它会影响到哪个系统，哪些 contract 不能破，适用的是哪些标准？

Which system does this touch, which contracts can't break, which standards apply?

M112:19

欢迎回到 AI Daily Brief。

Welcome back to the AI Daily Brief.

这周我有机会和 KPMG 一起去了 Utah，参加他们一年一度的 Tech and Innovation Symposium。

This week I had the chance to be out in Utah with KPMG for their annual Tech and Innovation Symposium.

这是我第二年参加这个活动了，而且每一次我都有机会做一场类似类型的分享。

Now, this is the second year that I've been at the event, and in each case have had a chance to do a similar type of presentation.

这一次，不管是我的重点，还是后面讨论的重点，都是想从更宏观的角度，总结一下当前正在塑造企业必须如何思考 AI 的那些大问题。

This time around, both my focus and the focus of the conversation after was about trying to sum up, broadly speaking, the big questions that are currently shaping how enterprises have to think about AI.

而让我觉得特别明显的是，和去年这个时候相比，这场讨论已经发生了多么大的变化。

And what's extremely notable to me is how much the conversation has changed since last year at this time.

接下来我会过一遍我当时那场分享的一个版本，不过在那之前，我其实想先把时间拉回到去年。

Now I'm going to go through a version of the presentation I gave, but before that, I actually want to zoom back to last year.

去年我讲的是 Where AI Is，15 Slides in 15 Minutes，虽然实际上，你也能看到，一共是二十三页。

Last year I did Where AI Is, 15 Slides in 15 Minutes, which was actually, as you can see, 23 slides.

现在回头看，我们当时在那个活动里讨论、并且觉得有意思或者很令人着迷的东西，几乎都有点天真了。

And looking back, it's almost quaint what we found interesting or fascinating in what we were discussing at that event.

第一个主题是 acceleration，我们当时谈的是，AI 不只是移动得更快，而是它被采用的速度本身也在变得越来越快。

The first theme was acceleration, and we talked about how AI wasn't just moving faster but was actually getting faster in the speed it was being adopted.

我当时提到，在五月到七月之间，Google 处理的每月 token 总量增长了百分之一百以上，达到了接近一千万亿个 token。

I discussed the more than 100% growth in the total monthly tokens that Google was processing between May and July, where they reached nearly a quadrillion tokens.

现在，你们很多人都知道，到今天这个阶段，一千万亿个 token，大概也就是一个没被盯着的 OpenAI 员工一个月能跑出来的量。但在当时，看到这么大的拐点，确实是件大事。

Now, as many of you know, a quadrillion tokens at this point is about what a single OpenAI employee left unattended will do in a month, but it was a big deal back then to see this massive inflection point.

而且实际上，那场分享里的一些主题，某种程度上也正是在为我们今天所处的位置做铺垫。

And indeed, some of the themes from that presentation were effectively setups to where we are now.

随着我们进入一个新时代，算力短缺不但没有缓解，反而变得更严重了。

The compute shortage has done nothing but get worse as we've moved to a new era.

当然，就算在那个时候，大家讨论得最热的话题也是 agents。

And of course, even back then, the big conversation was agents.

不过有意思的是，至少按我们今天使用 agents 的方式来看，agentic AI 在当时还很明确地属于未来的范畴。

Now, what was interesting is that, at least in the way that we use agents today, agentic AI was still firmly in the domain of the future.

那还是 Claude 3 Sonnet 和 O3 那个时间窗口，我们当时才刚刚开始理解那个以分钟为尺度的 task graph，它显示 AI 的能力每隔几个月就在翻倍。

This was the Claude 3 Sonnet/O3 type time horizon, and we were just wrapping our heads around the meter time horizon task graph that showed that AI capability was doubling every few months.

再说一个后来一直都很重要的主题，就算在那个时候，也已经很清楚，agentic coding 是最先爆发的 agentic use case。

Now, speaking of themes that would continue to be important, even back then, it was clear that agentic coding was the breakout agentic use case.

不过话说回来，为了让大家感受一下那到底是多久以前的事，当时我们所有人都震惊了，因为我们只用差不多一年时间，收入 run rate 就冲到十亿美元了。

Then again, to give a sense of just how long ago this was, we were all gobsmacked because we had hit a billion-dollar revenue run rate in just about a single year.

为了更明确地说明这变化有多大，就在这之前，我看到 Dwarkesh 发了一篇帖子，说 Anthropic 今年可能会做到一千亿到一千五百亿美元的 revenue run rate。

To put a fine point on how much this has changed, right before this, I read a post from Dwarkesh that suggested that Anthropic could get to $100 or $150 billion revenue run rate this year.

我就不把这些不同的幻灯片一页页讲过去了，但现在回头看，最让我印象深刻的是，当时关于 AI 和 agents 的问题，对不少人来说其实还停留在“会不会发生”的阶段。

Now, I won't go through all of these different slides, but what stands out to me reflecting back then was that the questions of AI and agents were really still for some if questions.

有一页幻灯片这张图里没有，但我记得在当时另一个更长的演讲里有，是 McKinsey 的一张图，显示的是已经落地至少一到两个、或者三种不同 AI use case 的组织数量增长情况。

One slide that I didn't have in this particular chart, but I know I had in a longer presentation that was being given around the same time, was this chart from McKinsey that showed the growth in the number of organizations that had implemented at least 1 or 2 or 3 different AI use cases.

对，在二零二五年年中的时候，一个很大的事仍然是，大概只有百分之四十的企业做到了两到三个 use case。

Yes, the big deal in mid-2025 was still that something like 40% of enterprises were up to 2 or 3 use cases.

一年之后，整个讨论已经发生了巨大的变化。

A year on, the conversation has changed immensely.

而我心里那个历史研究者的一面，会觉得很值得回头想想，我们到底是怎么走到今天这一步的。

And the historian in me thinks it's worth reflecting how we got here.

后来我们都知道，真正大的能力跃迁发生在那年年底，也就是十一月到十二月那段时间，我们迎来了 Opus 4.5 和 GPT-5.2。

The big capability jump, as we now know, came towards the end of the year, the November-December time period where we got Opus 4.5 and GPT-5.2.

不管具体原因是什么，就是在那几次模型更新里，agents 和 agentic workflows 真正开始大规模上线了。

For whatever set of reasons, those were the model updates where agents and agentic workflows actually came online in a major way.

不过很有意思的是，人们其实花了好几个月，才真正意识到有些东西已经变了。

Now, what was fascinating is that it actually took a couple of months for people to really grok that something had shifted.

大家都回家过节了，也稍微有点时间放松一下，等他们重新打开自己的 Claude Code，或者别的什么工具时，才发现自己能做的事情跟以前相比已经完全不一样了。

Everyone went home for the holiday, had a little bit of time to decompress, and when they fired up their instance of Claude Code or whatever tool they were using, they found that the stuff that they could do was significantly different than what it had been before.

我到现在还清楚记得，圣诞节到新年之间那一周，Twitter 上简直像海啸一样，创业者一个接一个，开发者一个接一个，满脸震惊地回来分享他们现在能做出什么，而那些东西以前根本做不到。

I still remember vividly the absolute tidal wave of tweets in that week between Christmas and New Year's of entrepreneur after entrepreneur and developer after developer coming back gobsmacked about what they could now build that they simply couldn't before.

更有意思的是，这几乎立刻也转化成了组织层面的实践。

Now, what's interesting is that this almost immediately translated into organizational practice as well.

这其中一部分原因是，软件组织在这一整年里，甚至更早之前，就一直在适应越来越强的能力。

Part of this was because software organizations had been adapting to greater and greater capabilities throughout the year and even before.

到了二零二六年初，软件工程组织早就不再把 AI coding 只看成一种 autocomplete 方案了，他们成了企业里最早一批真正完成转变的团队——不再把自己的工作理解成写代码，而是管理那些替他们写代码的 agents。

By the turn of 2026, we were long past software engineering organizations viewing AI coding as just an autocomplete solution, and they became some of the first groups in the enterprise to actually shift from viewing their job as writing code to managing the agents that wrote the code for them.

不过话说回来，也许是因为到那个时候，企业这边的人已经关注这件事超过两年了，所以并没有出现那种 AI 早期采用者和企业之间特别明显的时间差；企业也很快就在思考，这个全新的能力时代会给他们的工作带来什么。

That said, maybe because the enterprise folks had been paying attention for over 2 years at that point, it wasn't like there was some major lag from the AI early adopters to enterprises thinking about what this new age of capacity was going to mean for their work.

而进入二零二六年之后，绝对不只是软件工程组织在争分夺秒地把这些新的工作方式落到实践里。

And coming back into 2026, it was absolutely not just software engineering organizations that were racing to put into practice these new ways of working.

你会看到，从 marketing 到 legal，再到 finance，各个领域里的先锋建设者和早期采用者，也都开始琢磨怎么把这些新能力带进自己的工作里。

You saw vanguard builders and early adopters across domains from marketing to legal to finance starting to figure out how to bring these new capabilities into their work as well.

除了模型本身的跃迁，大家也意识到，新能力的一部分其实来自你把模型放进什么样的 harness 里。

Alongside the model jump, folks also recognized that part of the new capability set was actually about the harness that you situated the models in.

Claude Code 在整个二零二五年里一直都在不断扩大采用范围，但到了新的一年，它真正成了一个关注焦点，而这可能也被 OpenAI 全力押注他们的 Codex 产品进一步放大了。

Now, Claude Code had been growing in adoption throughout 2025, but became a real focal point in the new year, which was perhaps augmented by OpenAI going all in on their Codex product as well.

不过我觉得，从很多方面来说，真正让整个 harness 这个概念，以及说白了我们到底在讲 agents 是什么意思、构建和管理一个 agent 又意味着什么，这种理解大幅加深的时刻，是 OpenHands 火起来之后。

Still, I think in many ways where this whole idea of harnesses, and frankly, a much deepened understanding of what we actually mean when we say agents and what it means to build and manage an agent, came when OpenHands became popular.

几十万人，甚至如果把那些当时在中国排队等着拿到 OpenHands 使用资格的人也算进去的话，可能有上百万人，真的亲自动手，深入去摸清这些 agents 到底是怎么运作的。

Hundreds of thousands of people, perhaps millions if you include the people who were standing in line in China to get access to an OpenHands, really got their hands dirty figuring out the guts of how these agents work.

虽然现在你未必还会看到大家都在跑自己的 Mac mini setup，但我觉得，OpenHands 早期那段时间里爆发式的学习过程，未来会被看作 agentic AI 历史上的一个关键拐点。

And while you don't necessarily see everyone running their Mac mini setups anymore, the explosive learning of that early period of OpenHands I think will be seen as a key inflection point moment for the history of agentic AI.

当然，这一切并不只是发生在个体建设者身上，而某种根本性的变化已经发生的证据，也开始显现出来，尤其是在大实验室收入账面这一侧。

Now, of course, all of this wasn't just happening to individual builders, and the evidence that something fundamental shifted started showing up, particularly on the revenue side of the ledger for the big labs.

在今年头几个月里，感觉我们每一转身，Anthropic 尤其是它，都会放出一个让人下巴掉下来的新数字，讲他们的 revenue run rate 又涨了多少，最后甚至超过了 OpenAI——当然，这也不是说 OpenAI 自己的收入增长就慢。

For the first few months of this year, it seemed like every time we turned around, Anthropic in particular had released some new jaw-dropping number about how much their revenue run rate had grown, eventually eclipsing OpenAI, although it's not like they've been particularly slow in their revenue growth either.

现在，有意思的是，企业看到的是那张营收曲线的反面，也就是一张成本曲线。

Now, the interesting thing is that the enterprise experiences the inverse side of that revenue chart as a cost chart.

一方面，这其实一直都是不可避免的。

And on the one hand, this was always inevitable.

这些年来，我们一直在讲一个观点：企业里的 AI 不只是软件支出里的另一个分类，它代表的是某种根本上不一样的东西。

For years, we've been talking about the idea that AI in the enterprise is not just another category of software spend, but represented something fundamentally different.

它更像的，可能是劳动力。

Something more akin perhaps to labor.

那种体现在营收增长里的智能消费爆发，以及企业这边不断上涨的成本，本质上都只是这个事实开始真正显现出来的表现。

The explosion of intelligence consumption reflected in that growing revenue and the growing cost for enterprises were simply a manifestation of that fact coming to bear.

顺带一提，当大家意识到我们讨论的不是 seat，而是在讨论 token 之后，这也在很大程度上戳破了去年第四季度 Wall Street 上那些关于 AI 泡沫的说法。

Now, as an aside, the recognition that we were not talking about seats but instead talking about tokens did a whole lot to collapse the AI bubble narratives on Wall Street from Q4 of last year as well.

很快，我们就开始看到一些故事，说很多企业在短短几个月里就把全年预算烧得精光。

Pretty soon we were getting stories of enterprises absolutely torching their annual budgets in just a few short months.

其中最典型的就是 Uber，虽然这些故事被包装得很让人意外，但你真去想一想，其实一点也不奇怪。

Uber was the most notable of these, and although these stories were presented as surprising, if you actually think about it, it's really not that surprising at all.

当这些预算被制定出来的时候，根本没人知道 agentic token 时代的 AI 马上就要来了，那我们又怎么能指望组织提前把预算做得很到位呢？

How are we going to expect organizations to effectively budget for the agentic token era of AI when no one knew that that was right around the corner when those budgets were being made?

此后，常听这个节目的朋友也知道，过去这几个月一直主导讨论的就是这些主题：我们看到大家在朝这个新的 agentic 范式做适应，而且是朝着各种不同方向在演进。

Subsequently, and regular listeners of this show will know that these are the themes that have dominated for the past several months, we have seen adaptation to this new agentic paradigm run in all sorts of different directions.

有些地方我们看到的是 token cap，也就是公司开始设定每个用户每个月的使用上限。

In some corners, we're seeing token caps where companies are going with limits per user per month.

我们也看到，公司不得不去实验，去摸索怎么建立 measurement、monitoring 和 observability 系统。

We're seeing companies have to experiment with and try to figure out measurement and monitoring and observability systems.

随着成本不断飙升，这让一个原本在 harness 那场讨论里就已经出现的重点，被进一步放大了：我们讨论的早就不只是 AI 该选哪个 model，而是一个 architecture 和 systems design 的问题。

As costs spiral, it puts a whole new emphasis, something that was already coming up in the harness conversation, around the fact that we were no longer just talking about AI as a choice of which models, but as an architecture and systems design question.

当然，router——当下最火的产品——就是对此的一种回应。

The router, of course, the product du jour, is one response to this.

但是说到企业买家、规划者和战略制定者，我觉得没有人——至少我这周在 KPMG 活动上的交流已经很明确地证实了这一点——会把 OpenRouter 或者任何别的方案，当成解决所有这些问题的银弹。

But when it comes to enterprise buyers and planners and strategists, I don't think anyone, and certainly my conversations this week at the KPMG event have confirmed this, is looking to OpenRouter or any other solution as some silver bullet that's going to solve all these problems.

而且，新的问题还在出现。

And there are new problems.

具体来说，不管是在个人层面还是组织层面，capability gap 都在扩大。

Specifically, the capability gap is growing on both an individual and an organizational level.

所谓 capability gap，当然就是 AI 能做到的事情，和我们真正从里面拿到的价值之间的那段差距。

The capability gap, of course, is the space between what AI can do and the value that we're getting out of it.

好消息是，这个差距之所以变大，很大程度上是因为 AI 能力的上限正在以惊人的速度猛增。

Now, the good news is that it's grown largely because the upper bound of what AI can do is rocketing upwards at an incredible rate.

但即便如此，这个差距的扩大依然会带来非常真实的后果。

And yet still there are real consequences to that gap widening.

我一直特别强调的一个问题是，我相信 upskilling 这笔账很快就会以非常大的规模找上门来。

One of my bully pulpit issues is that I believe that the upskilling bill is coming due in a huge way.

当时如果 AI 学习还只是看你会不会写 prompt，也许你还能不在员工培训上投入太多。

When AI learning was just about whether you could prompt well, maybe you could get away with not investing a ton in training your workforce.

但现在，我们谈的是一种从根子上就不一样的工作基本单元。

Now, on the other hand, we are talking about a fundamentally new work primitive.

在很多学科和职能里，人们工作的方式正在发生核心性的变化：从“我自己做我的工作”，变成“我来管理替我工作的 agent”。

The way that people work is changing in a core way in many disciplines and functions from I do my work to I manage agents that do my work for me.

这就让培训需求相比上一代 AI 时代大幅提高了。

The need that that creates for training is radically heightened from the previous era of AI.

而且，确实，这次活动上很多人在讨论的一件事就是，怎么把这些本质上属于技术工具、而且能力极其强大的工具，分配给那些不是 engineer、也没有技术背景的人来使用。

And indeed, one of the things that a lot of folks are talking about here at this event is how to deal with apportioning these incredibly powerful tools that are inherently technical tools to folks that aren't engineers and aren't technical by background.

这次活动上流传着很多故事，说有人不小心把 agent 放到了关键系统里，不一定是因为他们做错了什么，而是因为缺少合适的 guardrail 或 access provisioning，再加上这些能力极强、而且现在又更有韧劲的 model，并不会老老实实待在自己的边界里。

There are a lot of stories floating around this event of people accidentally unleashing agents on critical systems, not because even necessarily they were doing anything wrong, but because there weren't the right guardrails or access provisioning, and these incredibly capable models with their new tenacity just didn't stay in their boxes.

这不只是一个 upskilling 的问题。

Now, this is not an upskilling question alone.

还是那句话，当下的关键词是 systems 和 architectures。

Again, the watchword of the moment is systems and architectures.

但是如果没有这些培训，组织几乎注定会越来越频繁地遇到这类问题；反过来说，也可能会因为不信任员工，而限制那些本来完全可以用这些工具做出巨大价值工作的人去真正使用它们。

But without that training, organizations are almost doomed to face this sort of issue in increasing fashion, or on the other hand, restrict the opportunity for people who could really be doing incredibly valuable work with these tools to do so because they're not trusted to do so.

这就把我们带到了这些问题上——这些问题不只是这场演讲后面的 panel discussion 里在讨论，说实话，整个活动现场各种边上的聊天里也都在反复聊。

And this gets us to the questions that were explored not only in the panel discussion that followed this presentation, but honestly in these side conversations all over the event as well.

第一个问题是，企业要怎么为 agentic 时代重新设计自己？

The first question is, how are enterprises redesigning for the agentic era?

这里的关键词是，重新设计。

And the key word here is redesigning.

像 panel 里的 KPMG 的 Steve Chase 这些人，给出的最大提醒，就是别想着只是把一个 AI strategy 硬接到现有流程和系统上，因为这么做会带来很多问题和负面影响。

The biggest caution that folks like Steve Chase from KPMG on the panel had was the warning of the problems with and ill effects of trying to simply bolt on an AI strategy to existing processes and systems.

其实这一直都是个问题，而且至少也会让 AI 的潜力发挥不出来，就算是在我们还明显处在 assisted AI 和 efficiency AI 时代的时候也是这样。

Now, that has always been problematic and at least under-maximizing for the potential of AI, even when we were firmly in the assisted AI and efficiency AI era.

但到了现在这个 agentic 能力兴起的阶段，这种问题就更严重了。

But in this time of new agentic capability, that gets even worse.

相关的第二个问题，是这种重新设计本身到底是什么性质，以及为什么组织需要从 architectures、systems 的角度去思考，而不只是盯着 models。

Relatedly, the second question is about the nature of that redesign and why organizations need to be thinking in terms of architectures, systems, not just models.

如果以前一个组织面对技术带来的新挑战时，做法是先看看哪个 vendor 最适合解决这个问题，那放到我们现在所处的这个阶段，这就已经远远不够了。

If previously an organization's response to some new challenge wrought by technology was to figure out which vendor was best suited to solving that problem, that is simply insufficient for the moment that we find ourselves in now.

思考 architectures，意味着要去思考复杂的 model systems，让不同类型的任务可以调用不同层级的智能。

Thinking about architectures means thinking about complex model systems that allow different levels of intelligence for different types of tasks.

这也意味着要去思考，没错，就是 routing systems——不管是现成买来的产品、bespoke 的，还是别的什么方案——来让这种路由真正发生。

It means thinking about, yes, the routing systems, whether they are products off the shelf or bespoke or something else that allow that routing to happen.

但这也跟 harness design 有关，也就是哪些职能、哪些人，可以访问什么类型的 context、data 和 systems integration，以及围绕这些东西需要设置什么样的 guardrails。

But it's also about that harness design, about which functions and people have access to what types of context and data and systems integration and what the guardrails that surround it need to be.

再到第三个问题，就是你要怎么把成本分配到不同的群组之间。

And as we get into the third question, how are you provisioning costs across different groups?

而这背后最核心的东西，是另一种系统设计上的需求，也就是要有监控和衡量 AI usage 的系统。

The big thing that underlies that is another systems design need, which is systems for monitoring and measuring AI usage.

哥们，我感觉自从 crypto 最火的时候以后，我就没在别的活动上听过 token 这个词被提到这么多次。

You have not seen the word token used more at an event since the height of the crypto era, man.

当然了，这次活动里说的 tokens 显然跟当时说的很不一样，但这里大家都有一个非常广泛的共识：如果你对 AI 的成本，以及它和产出之间的关系，没有更好的可见性，那你就会很难判断，到底哪些个人、哪些团队、哪些职能、哪些项目，应该获得哪些类型的 models 的访问权限，以及多大规模的访问。

And obviously the tokens we're talking about at this event are very different, but there is a very broad recognition here that without better visibility into the cost of AI and its relationship with outputs, it gets very hard to figure out which individuals, which groups, which functions, which projects should be getting access to which types of models and at what magnitude.

考虑到我前面提过的那个 bully pulpit，我确实很欣慰地看到，这些组织其实都把 enablement 和 education 当成了一个非常大的关注点。

Given that bully pulpit I mentioned before, I have certainly been gratified to see how big a concern enablement and education really is among these organizations.

如果要我概括一下我看到的主流讨论，大概就是很多人都摊手说，算了，我们只能自己来做了，于是开始尝试各种 bespoke 的定制化方案，去做出适合自己组织、也适合自己人群的办法。

If I had to characterize the average discourse I've seen around that, there is a lot of throwing up of the hands and saying, screw it, we're just going to have to do this ourselves, and experimentation with bespoke customized solutions for this that work for the organization and the population that it has.

换句话说，大家已经意识到，这不会是那种像很多年前企业培训一样，靠一堆花里胡哨的视频课程就能搞定的事；相反，这会是很真实、很混乱、很费劲的工作——要让人们用新的方式使用这些工具，去做以前没做过的新事情，然后再想办法把知识从那些已经摸索得比较好的组织部门，传递给那些还没摸索明白的部门。

In other words, there's a recognition that this is not going to be a bunch of cute video courses of the pattern of corporate trainings of yore, but instead is going to involve the real messy work of getting people to use these tools in new ways to do new things, and then figure out how to transmit knowledge between parts of the organization that are figuring it out well versus parts that are not figuring it out so well.

确实，我一遍又一遍看到的一个大模式，就是各种形式的协作：一方面是经过 AI 重新设计的软件工程组织和业务单元之间的协作，另一方面也是 AI 早期采用者、AI champions 和其他类型业务单元之间的协作。

Indeed, one of the big patterns that I am seeing over and over and over again is various forms of collaboration between both AI-redesigned software engineering organizations and business units, but also AI early adopters and AI champions and other types of business units.

我觉得现在这类对话更成熟的地方在于，已经没人会说让 marketing 的人去替代 engineers 了；但大家现在会讨论的是，engineers 或 product managers 身上大概有百分之十到百分之二十那类技能，甚至比技能更重要的 mindset，可以变成其他职能的人——不管是 marketing、sales、back office 还是什么——手里那套核心工具箱的一部分。

I think the sophistication in the conversation is that no one is talking about the marketing folks replacing the engineers, but they are now talking about the 10 or 20% of the types of skills, and even more than that, mindsets that engineers or product managers have that can become a part of the essential toolkit for those people in other functions, be it marketing or sales or back office or what have you.

以及，怎么才能把这种新的传递方式做到最好。

And how to best do that new sort of transmission.

我会说，这次活动上的很多讨论，重点都放在了内部转型上。

Now, I would say that a lot of the discourse at this event has been focused on internal transformation.

对这批具有代表性的企业样本来说，二零二六年非常明确地会是这样一年：AI 不是一个 technology problem，而是一个 transformation problem——这件事已经真正变成了必须面对的现实。

2026 is very clearly the year that, for this representative sample of enterprises, AI is not a technology problem, but a transformation problem has really come home to roost as the reality.

但与此同时，agentic transformation 其实还有完整的另一层维度，那就是外部会发生什么。

And yet, there is also the entire dimension of agentic transformation that has to do with what happens externally as well.

换句话说，就是 agentic 带来的机会，正在怎样重塑 business cases。

In other words, how are agentic opportunities reshaping business cases.

这里大家在讨论的一些例子，包括 business model 的转变，比如有人在尝试用 outcomes-based pricing，来替代像按小时收费这种 input-based pricing。

Some of the examples of that that people are discussing here include shifts to the business model, people experimenting with outcomes-based pricing instead of input-based pricing like hourly billing.

也有人在讨论，在这个新的环境里，会出现哪些新的产品类型和新的服务类型。

There is some discussion of new types of products and new types of services that become available in this new context.

还有很多人也在重新评估，老产品原本那个核心形态，到底还意味着什么。

And there's also a lot of reevaluation of what the core state of the old product actually means.

比如说，如果很多审计工作都可以由 agents 来做，而且还不是一次性的做，而是持续不断地做，那 audit 到底还算是什么？

What is, for example, an audit if agents can be doing a lot of that work and if they can be doing it not just on a one-off basis, but on a persistent basis?

我的感觉是，虽然这种类型的讨论确实在发生，但大多数组织都把自己看成是——姑且这么说吧——自己外部 AI strategy 的 patient zero，所以他们现在更专注的是先把自己的工作方式补强、稳住，而不是马上就对外销售的东西做非常激进的改变。

It feels to me as though that while that type of conversation is happening, most organizations are viewing themselves as patient zero, let's call it, for whatever their external AI strategy is, and are focusing on shoring up how they work first before necessarily making radical changes to what they sell externally.

不过，当然，对某些类型的组织来说，这种变化是被迫发生在他们身上的。

Although certainly for certain types of organizations, that change is being forced upon them.

当然，说到商业模式被颠覆这件事，情况之所以更难，是因为没有人可以直接把一切停掉六个月，专门把这事琢磨明白。

Now, of course, when it comes to business model disruption, it's made all the more difficult by the fact that no one gets to just shut things down for 6 months to figure this all out.

他们得实时去做这件事，哪怕同时还在用老产品、老交付方式，服务那些老客户。

They gotta do it in real time, even as they're servicing legacy customers on legacy products with legacy methods of delivery.

除此之外，我们讨论过、而且现场一直在萦绕的最后一个问题是：如果、也当我们成功设计出新系统时，我们怎么把一种动态性建进去，让它几乎自带一种“计划性淘汰”，并且把对短暂性的理解也内置进去？

And on top of all of this, the last question that we explored and that was floating around here is if and as we are successful in designing new systems, how can we build dynamism into that that has almost planned obsolescence and an appreciation of ephemerality built into it?

围绕它们的那些框架会变。

The harnesses around them are going to change.

交互模式会变。

Interaction patterns are going to change.

客户预期会变，市场预期会变，政策也会变，所以无论造出什么新东西，都必须预先假设、并且按这样一个事实来设计：从它准备好的那一刻起，往后几个月，很可能又得再改一遍。

Customer expectations are going to change, market expectations are going to change, policy is going to change, and so whatever new thing gets built has to assume and design for the fact that a few months down the line from whenever it is ready will likely require it to change all over again.

如果这一切听起来让人头都大了，确实是这样，但我觉得这里面也有非常积极的一面。

If all of this sounds head-spinning, it is, but I think that there is something immensely positive.

去年，哪怕就在这个活动上——这已经算是企业活动里最 AI-pilled 的那种了——还是像我说的那样，充满了各种“如果”的问题。

Last year, even at this event, which is about as AI-pilled as an enterprise event can be, there were still, as I said, so many if questions.

我该怎么说服组织里的其他人，让他们相信这是真的，而且我们应该去做？

How do I convince others in my organization that this is real and that we should be doing it?

我该怎么展示 ROI，来证明我们在做的事值得投入这些时间和金钱？

How do I show ROI to prove that what we're doing is worth the time and money that we're spending on it?

现在倒不是说 ROI 这类问题消失了，但总的来说，大家现在在问的问题，在我看来，更像是为一个新时代做重新设计时必须回答的基础性问题，而且这些问题，接下来大概五年里，我们都会一直在回答。

Now, it's not that ROI questions and things of the like are gone, but by and large, the questions that people are asking now are, it feels like to me, the foundational questions for redesigning for a new era that we are going to be answering for the next, call it, half decade.

那些开始讨论怎么设计和分配 token 预算的公司，现在正在探索一种新的支出类别，而这很快就会变成他们组织里不可或缺的一部分。

Companies asking about designing and allocating token budgets are now exploring this new category of spend that is just going to become an essential part of their organization.

当公司在讨论要围绕他们正在使用的新智能去构建 observability 系统时，虽然模型和外围框架可能会变，但很可能无论之后更新成什么，依然都需要这种 observability。

When companies are talking about building observability systems around the new intelligence they're using, while the models and harnesses may change, it is very likely that whatever gets updated is still going to need that sort of observability.

我想重点就是，这次范式转变已经发生了。

I guess the point is that the paradigm shift has happened.

这些年来，基本上从 ChatGPT 横空出世那个时刻开始，企业一直都在期待从 assisted AI 转向 agentic AI，也就是 AI 不只是帮我们做工作，而是真的自己去把工作做掉。

For years, basically since the ChatGPT moment, enterprises have been anticipating the shift from assisted AI to agentic AI, the opportunity for AI not just to help us do work, but to actually do the work itself.

现在这已经到来了，所有问题都变成了：我们该怎么解决这种新工作方式带来的全新问题，以及怎么最好地抓住它开启的那些机会。

Now that that is here, all of the questions are about how we solve all the new problems that that new way of working brings and how we best seize the opportunities that it opens up.

现在几乎没有哪个问题已经有答案了，但我觉得，至少大家现在问的是对的问题，这本身就挺让人振奋的。

Almost none of the questions have answers right now, but it should feel good, I think, that the questions being asked are the right ones.

总之，感谢 KPMG 邀请我来。

Anyways, thanks to KPMG for having me out.

这是个很棒的活动，我也很期待明年再回来。

It was a great event, and I look forward to coming back next year.

说实话，到那时候会变得有多不一样，我现在都想象不到。

Where honestly, I can't even imagine how different it's going to be by then.

好了，今天的 AI Daily Brief 就到这里。

For now, that's going to do it for today's AI Daily Brief.

一如既往，感谢你的收听或者收看。

Appreciate you listening or watching as always.

那我们下次见，peace。

And until next time, peace.

已剔除 3 处广告（点击展开查看）

M10:12广告 · 已剔除

The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI.

Alright friends, quick announcements before we dive in.

First of all, thank you to today's sponsors, KPMG, Blitzy, Retool, and Airtable.

To get an ad-free version of the show, go to patreon.com/aidailybrief, or you can subscribe on Apple Podcasts.

To learn more about sponsoring the show, send us a note at [email protected].

And to bone up on your AI skills with the last month or so of summer, go check out our latest free self-directed education program, this one being a choose-your-own summer adventure.

You can find it at summeradventure.ai.

M19:32广告 · 已剔除

KPMG and the University of Texas at Austin just analyzed 1.4 million real workplace AI interactions and found something surprising: the highest-impact users aren't better prompt engineers—they treat AI like a reasoning partner.

They frame problems, guide thinking, iterate, and push for better answers.

And the good news?

These behaviors are teachable at scale.

If you're trying to move from AI access to real capability, KPMG's research on sophisticated AI collaboration is worth your time.

Learn more at kpmg.com/us/sophisticated.

That's kpmg.com/us/sophisticated.

Blitzy's deep codebase understanding unlocks the thing every roadmap owner cares about: shipping new features.

M110:23广告 · 已剔除

Blitzy already knows because it reverse-engineered your entire codebase into a dynamic knowledge graph before feature work began.

With that complete picture, Blitzy builds features end-to-end— architecture, APIs, UI, and tests all validated against your existing systems.

One Blitzy customer built an AI-native application from scratch with 100% autonomous completion, saving over 2,700 engineering hours.

Features that respect your codebase instead of fighting it.

Stop letting your backlog grow faster than your team.

Accelerate your roadmap at blitzy.com.

That's B-L-I-T-Z-Y dot com.

This episode is supported by Retool.

AI made building software easier than ever, so more people are building it than ever, usually without a thought for security.

Right now, people in your company are vibecoding, and every ungoverned app that touches your data is a risk you own.

Retool takes that risk off your shoulders.

Build apps however you want— natively with Retool, or with Claude Code, Codex, or any coding agent— and ship it in a secure, governed environment.

Security lives in the platform, not in each app, so however it was built, it's governed the moment it ships.

It's why teams at Amazon, Stripe, and Brex build on Retool.

And new enterprise customers who sign up by September 30th get up to $10,000 in AI credits per year.

Learn more at retool.com/AIDaily.

This episode of the AI Daily Brief is brought to you by HyperAgent, where you run fleets of agents your team can manage together.

New users get $1,000 in inference.

Forget local agents and chat workflows waiting on your laptop to be prompted.

HyperAgent deploys always-on agents in the cloud, doing real work across the tools your team already uses.

Marketing's agent turns competitor moves into landing pages.

Sales' agent enriches leads, drafts emails, and updates the CRM.

Ops' agent chases the paperwork and tracks the budget.

Every agent has access to shared context and follows your rules about scope and approvals.

It's time you add agents that feel like teammates.

Hire yours at HyperAgent, built by the team at Airtable.

Claim your $1,000 in inference at hyperagent.com/AIDailyBrief.