OpenAI对话增强器：深度解析

你们订阅了 aichatdaily.com 吗？

Are you guys subscribed to aichatdaily.com?

是的。

Yeah.

我每天都会发一封简报，内容跟播客里讲的一模一样，都是那些故事，只不过我把它拆成了特别容易消化的格式，还附上相关文章的链接，方便你深入了解。

I have a newsletter that I send out every single day, and it's essentially all of the same content, all the same stories that I talk about in the podcast, but I do them in these really easy to break down formats, and I also have links to articles to go more in-depth.

这就像播客的节目笔记，但内容要深入得多，还有一些我们在播客里没时间讲的额外故事。

So it's kind of like show notes for the podcast, but it goes into way more depth, and we have extra stories that we can't get around to covering in the podcast.

所以，如果你每天想在开始或结束一天的时候，快速浏览一遍最新的AI新闻，就去 aichatdaily.com。

So if every single day you wanna start the day out or end the day by getting a roundup of the latest in AI news in a really easy to scan format, go to aichatdaily.com.

页面右上角有个明显的订阅按钮。

There's a big subscribe button in the top corner.

我可能会在节目笔记里放个简报的链接，还有那个内容丰富的完整新闻网站。

I'll probably leave a link in the show notes to the newsletter, and also just a full news site with tons of articles.

我每天大概会发布十篇左右的文章。

I'll probably publish about 10 of them every single day.

所以如果你想去看看，网址是 aichatdaily.com。

So if you wanna go check that out, aichatdaily.com.

好的。

Okay.

OpenAI 推出了 GPT Live One。

OpenAI is shipping GPT Live One.

这是一个全双工语音模型。

So this is a full duplex voice model.

全双工语音模型是什么意思呢？这其实挺让人惊讶的，因为以前在 OpenAI 工作的 Miriam Maradi，最近在她新创立的 Thinking Machine Labs 发布了一个类似的东西，结果一看，OpenAI 这分明是直接抄了她的方案。

And what duplex voice model means well, this is actually kind of crazy because, Miriam Maradi, who is formerly at OpenAI, she released something recently with her most recent company, Thinking Machine Labs, where it's I mean, essentially, it looks like OpenAI has just knocked her off.

这事儿挺有意思的。

So this is gonna be interesting.

另外，中国已经对 Anthropic 的 Claude Code 发出了警告。

Also, China has flagged Anthropics Claude Code.

他们说这存在安全风险，因为可能有后门问题，不过具体怎么样，我也说不准。

They say that this is a security risk because there's some backdoor risks with it, which, I don't know.

我觉得，中国说任何东西有安全风险，这本身就挺讽刺的。

I just think anything with security risk is ironic coming from China.

NVIDIA 的 Nemotron 3 Ultra 刚刚以十分之一的成本达到了闭源模型的同等水平，这是我最近在 Lane Chain 上做实验得出的结果。

NVIDIA's Nemotron three Ultra has just hit closed model parity at a 10 x lower cost, and this is from some experiments I've been doing on lane chain.

每当我们能把这些模型的成本降下来，我都特别兴奋。

I'm really excited when we're able to get the costs down on some of these models.

Meta 在他们的 AI 眼镜上加了防篡改的 LED 灯，还从眼镜其他部分收集更多数据，所以那边的情况挺复杂的。

Meta is adding a tamper proof LED to their AI Meta glasses, and they're also getting more data collection from other parts of the glasses, so there's kind of a mixed bag with what's going on there.

OpenAI 的首席未来学家约翰·阿切姆要离职了。

An AI the OpenAI chief futurist, John Achyem, is leaving.

他在那儿待了九年。

He's been there for nine years.

他是 OpenAI 最早一批负责信任与安全的元老级人物。

He's one of the OG trust and safety people over OpenAI.

咱们还是先聊聊 OpenAI 最新发布的模型 GPT Live One 吧。

But let's kick the podcast off with OpenAI's latest model, GPT Live One.

这是一个语音模型。

This is a voice model.

它能同时听和说，这一点你可能已经注意到，以前很多语音模型都很难做到——比如你现在用的ChatGPT语音模式，而这个新模型其实取代了他们之前的高级语音模式。

It listens and it speaks at the same time, which is something that you've probably noticed a lot of these voice models have struggled with in the past if you're talking to, you know, what voice mode is on ChatGPT right now, which by the way, this replaces their advanced voice mode.

如果你以前用过，就会知道，你讲完之后得等几秒钟，它才回应你。

But if you if you've ever talked to it before, you talk, there's a couple seconds of latency, then it responds back to you.

他们花了很多功夫，尽量把延迟降到最低。

And they've done a lot of work to try to like get that latency down as much as possible.

但这个新推出的Live One模型真正厉害的地方在于，它能真正实时倾听，同时听和说，因为这正是人类对话时的样子。

But what's really cool about this new Live One model is that it's literally listening, and it's able to listen and talk at the same time because that's how actual humans talk when we're in a conversation.

我们经常会打断对方。

We kinda interrupt each other.

也会插话抢着说。

We talk over each other.

对方正说着话呢。

They're saying something.

你在说话。

You're saying something.

我的意思是，我们并不是故意要打断别人、抢话，但这本来就是交流的一部分。

And I mean, the goal is not to interrupt and talk over each other, but it's just part of communication.

对吧？

Right?

我们并不会非得等别人说完、有个完美的停顿才开口。

Like, we don't always wait for there to be a perfect pause at the end of everything that someone says before we jump in.

所以这些模型现在在这方面越来越擅长了，这真的很有趣。

And so that's what these models are getting really good at, which is really interesting.

对我来说特别惊人的是，几个月前米里亚姆·穆拉迪就已经发布过类似的模型了。

And what's, to me, really crazy about this is that Miriam Muradi released a model just like this a number of months ago.

我记得在播客里专门聊过这个。

I remember covering it on the podcast.

那是Thinking Machine Labs的主打产品，我当时就觉得，这想法真棒。

And it was like the big thing from Thinking Machine Labs, and I was like, man, this is a really cool idea and concept.

结果呢，过几个月后，只要有人推出什么新功能，Frontier Labs 马上就能复制出来，逆向搞懂，然后自己也上线，毕竟他们不想被甩在后面。

Well, turns out, you know, a couple months later, anyone that releases any sort of cool new feature, a couple months later, the Frontier Labs are gonna be able to clone it, reverse engineer, and put it out themselves because, you know, they don't wanna fall behind.

我觉得这个功能特别重要，是因为它把三件事合到一个步骤里了。

The reason why I think this is one in particular is an important feature is because there's three separate things that are getting collapsed into one step.

第一，它大幅减少了延迟，能让对话变得自然得多。

Number one, it's cutting the delay, so it's gonna let your conversation be way more natural.

而且他们还要把这个功能推给一亿五千万使用 ChatGPT 语音功能的用户。

And they're also gonna be rolling this out to a 150,000,000 ChatGPT voice users.

最后，这个机制本身特别有意思。

And finally, the way this works is really interesting to me.

有个叫 GPT Live One Mini 的东西，主要给所有免费用户用。

So there's something called GPT Live One Mini, and this is gonna be for basically all of the free users.

这算是默认配置了。

It's kind of the default.

但如果你是付费用户，每月花二十美元，就能用到更强大的 GPT Live One，它会在对话中把一些难的问题转给 GPT 5.5 处理。

But if you're on the paid tier, if you're paying $20 a month for ChatGPT, you get the larger GPT Live One, which is gonna send some of the harder questions over to GPT 5.5 mid conversation.

所以当你在对话中说一些比较复杂的内容时，它会中途截取你一半的句子，发给GPT 5.5生成回复，然后在你还没说完之前就把答案吐出来，这简直太惊人了。

So you're gonna be talking in the middle of the conversation if you're saying something kind of complex, and it it's gonna, like and basically get half your sentence, send it over to GPT 5.5 to try to come up with a response and send it back and put it out before you even finish your sentence, which is really crazy.

它甚至能在你说话中途打断你。

It can be interrupted in the middle of a sentence.

如果它在说，你随时可以插话打断它。

So if it's talking, you can interrupt it.

它能处理你思考时的停顿。

It handles silences while you think.

你只是稍微停顿一下想一想，它就不会急着插话，这一点真的很好。

So just because you're pausing to think about something doesn't mean it feels like it has to jump in, is something really annoying.

我以前都觉得，自己根本来不及整理思路。

I feel like I can barely like keep my thoughts.

跟这些系统说话时，我连思考都困难，因为一停顿，它就马上抢着接话，逼得我只能说个不停。

Like I can barely think when I'm talking to these things because I gotta talk so fast because if I stop talking, it's gonna wanna jump in and start talking to me.

所以我特别欣赏这一点。

So I really appreciate that.

它还会在你讨论的内容旁边显示图表或图片，以前呢，屏幕上就只是个不停闪烁的动画。

And it's gonna show charts or images alongside whatever you're talking to it about, which, you know, before it was just kind of this like animated strobe on your screen.

但现在，如果你边走边问，比如‘最近天气怎么样？’

But now, if you're talking and you're like, hey, know, what are the, you know, what's the weather gonna be like?

它一边跟你说话，一边就在屏幕上调出天气图表，视觉加听觉双重信息，真的很棒。

It's pulling up weather charts on the screen while it talks to you, so you get visual and audio, which is really cool.

负责这个产品的负责人是阿迪·阿莱蒂。

The product lead on this is or the product lead is Adi Aletti.

他说自己曾经在散步时，跟这个模型聊了三十到四十分钟，我觉得他们特意提到这一点，是想说明这玩意儿特别适合处理长时间的任务。

He said that he held a thirty to forty minute conversation with the model on walks, and I think the goal, and probably the reason why they're telling us this, is because they're trying to make this something for, or trying to explain this is really good for long running tasks.

这可不是那种简单的指令，比如‘今天天气咋样？’

This isn't just quick commands like, hey, what's the weather like?

而是像这样：‘我想好好梳理一下我打算做的这个商业计划，你陪我聊聊吧。’

It's like, hey, I'm trying to break down this complex business plan that I wanna, you know, create.

跟我一起讨论一下这个。

Talk with me about this.

或者我想弄清楚XYZ的最佳策略是什么，你跟我聊聊这个。

Or I'm trying to figure out what the best strategy is for XYZ, talk to me about this.

然后你花上三四十分钟，真正地去学习、规划，或者把事情办成。

And you go and have a forty minute conversation to learn something, to plan something, to actually get something done.

说实话，我之前一度疏远了OpenAI和ChatGPT，因为感觉它们并不专注这方面，所以我转投了Claude。

Which if I'm being honest, I was kind of moving away from OpenAI and Chat GPT for a while because it didn't feel like that was their focus, and and I was going to Claude.

但最近我一直在试用ChatGPT，主要是当我在Claude那边用完令牌限额时，就又回过头来用它。

But I've been trying out a bunch of things on ChatGPT lately, mostly like whenever my, whenever my tokens, my token limits reach on Claude, I go back to ChatGPT.

我对它能做的研究印象很深，比如我说：‘你去抓取这五千个数据，分析一下这些网站或产品，然后给我整理出一份清单，列出三百六十五个我能做的XYZ相关事项。’

And I've been really impressed by the research it can do, where I'm like, hey, you know, go scrape like these 5,000 things and do an analysis on these different websites or these different products, and like bring me back, know, help me create a list of three sixty five x y z things I can do for this.

它真能去干点实在的活儿。

Like, it goes and gets some real serious work done.

所以我现在特别兴奋。

So, I'm really excited.

我觉得这之所以特别重要，不只是因为对人有用，还因为OpenAI目前正在开发大量与耳机和可穿戴设备相关的产品。

I think the reason why this is particularly important beyond just humans is because we're gonna have a lot of work with earbuds and wearables that OpenAI is currently working on.

所以他们正在开发的这些硬件，这些工具将会非常有用。一旦谈到机器人、通信这些事，沟通的流畅性就变得至关重要。

And so, a lot of the hardware that they're making, these tools are gonna be really useful for, and, you know, when we start talking about robots getting them and all that kind of stuff and being able to communicate with that, this is gonna be important that the communication is as smooth as possible.

中国把Anthropic的Claude Code列为安全后门风险，这简直可笑，因为多年来中国自己就被指责在服务器和各种硬件里植入后门。

China is flagging Anthropix Claude Code as a security backdoor risk, which is hilarious because China's been accused of putting backdoors into chips and hardware inside of, China for for many years on, you know, servers and all sorts of crazy stuff.

不过这次，我觉得他们的说法很可能确实有道理。

But this in particular, I do think is probably accurate.

他们声称Claude Code存在一个后门漏洞，会泄露你的位置和身份信息，并警告所有在中国的用户，如果安装的是四月到六月之间的版本，必须立即升级或卸载。

What they've said is that Claude Code has a backdoor vulnerability that's gonna leak your location and your identity data, and they're telling everyone in China to uninstall or to upgrade the version, if you have something installed between April and June to upgrade it.

这个警告挺有意思，因为Claude Code根本没在中国正式销售——阿里巴巴早就全面禁用了Anthropic的所有工具。

The advisory is interesting because Claude Code isn't officially sold in China, so, you know, Alibaba has banned Anthropic tools company wide.

这事儿发生在Anthropic指控阿里巴巴窃取其AI技术的整整一周后。

This is also a week after Anthropic accused Alibaba of stealing their AI technology.

让我觉得特别有意思的是，有迹象表明，Exomi内部的一些人曾在三月份的一个国家主办的论坛上，公开承认他们在使用Claude Code，尽管这工具在中国官方是禁用的。

And what was interesting to me is there's like some moments where different people inside of, I think, Exomi, they had like an AI developer conference, I think, and they were kind of publicly acknowledging that they were using Claude Code at a state organized forum, this was back in March, even though it's officially not available in China.

所以很讽刺：官方说不提供、不销售，阿里也全面禁止，没人该用，可中国却在提醒所有人：我们知道你们在偷偷用。

So it's funny because it's like officially not available, officially not sold, Alibaba's banning it, no one's supposed to be using it, and yet China's like warning everybody like they know they're using it.

总之，这事儿对我来说挺离谱的。

Anyways, it's kinda crazy for me.

不过我觉得重点是，他们提到的漏洞版本是2.191到2.19656，而我们现在用的已经是2.12了。

What I do think is important though is this particular flag they have is version 2.191 to 2.19656, but we're already on 2.12 for the version that we're currently on.

所以如果你用的是最新版的Anthropic产品，这事儿跟你关系不大。

So if you're on the latest Anthropic thing, this isn't really relevant to you.

现在大多数现代编程助手，比如OpenAI和Google的，都会把遥测数据和崩溃报告发回他们的服务器。

Most modern coding assistants, so if you look at like OpenAI and Google, they're selling, they're sending what's called telemetry and crash reports back to the main servers.

每次出问题，我觉得这才是中国担心的——崩溃报告里包含的用户信息太多了。

And anytime that something happens, I think this is what China's worried about, the crash reports they're worried has too much information about the users in it.

但中国并没有说清楚，为什么只针对Claude Code，而不提OpenAI和Google。

But China wasn't really clarifying if Claude Code was the only one doing this, why they were singled out and not like OpenAI and Google.

我都不确定Google的产品是不是真能在大陆用。

I'm not even sure if Google's is available in China.

总之，中国这地方真是让人摸不着头脑。

So anyways, it's China's a crazy, crazy place.

对于开源模型来说，这是个重大突破：NVIDIA 的开源 Nemotron 3 Ultra 模型在商业任务上的表现已经能和顶尖的闭源 AI 系统媲美，当然不是所有任务，而是专门针对商业场景，而且成本只有十分之一。

In a big win for open source models, NVIDIA's open source Nemotron 3 Ultra model now matches the best closed AI systems on business tasks, so not everything, but business tasks specifically, and it can do them at a tenth of the cost.

LangChain 没有重新训练模型，而是围绕它优化了代理软件的配置，证明了只要工程环境做得好，就能在企业级任务上抗衡那些昂贵的专有 AI。

LangChain actually tuned the agent software around the model without retraining it, and they basically proved that the right engineering environment can compete with really expensive proprietary AI on when you, when it comes to doing like enterprise work.

所以，只要你对这些开源模型做点微调，就能以远低于 Anthropic 或 OpenAI 的成本，达到差不多的效果。

So, you can get these open source models, if you fine tune them, you can get them just as good for way cheaper than something like Anthropic or OpenAI.

他们调整了系统提示词、工具描述和中间件，把性能压到极致，然后拿去跑了一轮深度代理基准测试。

Lane chain adjusted system prompts, tool descriptions, and middleware, they did all of that to really just squeeze the top tier performance, then they put it through the deep agent benchmark.

Ambridge、Amdocs、Box 这些公司已经在用这套技术栈为自己的平台和工作流开发专用代理了。

Ambridge, Amdocs, Box, and are already building specialized agents on this exact stack for their platforms and their workflows.

这个优化后的模型已经上线六个云平台，你可以在 Bastion、Cruso Cloud、DeepInfra、Firework、Nebius 和 Together AI 上用到它——我用的就是 Together AI 来跑 AI Box。

The tuned model is live on six hosting providers, so you can get it on Bastion, Cruso Cloud, DeepInfra, Firework, Nebius, and Together AI, which is what I use for AI Box.

所以，现在想用的话，其实挺容易找到的。

So it's, you know, pretty widely available.

这真的让人很兴奋，我敢说 AI Box 上现在就已经部署了 Nemotron 模型。顺便说一句，如果你还没试过 aibox.ai——这是我自己的创业项目，上面已经集成了八十多个不同的 AI 模型。

This is really exciting, and I believe we actually do have Nemotron over on AI Box right now, which by the way, if you haven't tried aibox.ai, my own startup, we have over 80 different AI models all on there.

每月只需8.99美元，你就能访问所有顶尖的AI、图像、音频、视频，甚至音乐模型。

So for $8.99 a month, you get access to all of the top AI, image, audio, video, and even music models.

如果你不想到处订阅各种服务，还要在这么多平台登录账号，那就直接去一个地方：aibox.ai。

So if you don't wanna have to have subscriptions to all of these different things and have, you know, logins on a ton of different platforms, just go to one place, aibox.ai.

我们还提供了一个MCP，能把这些模型和Claude这类工具无缝连接起来。

We even have an MCP that will tie all of those models into things like Claude.

如果你想在Claude里使用图像、音频和视频功能，用我们的MCP就行，每月8.99美元。

So if you wanna get like image and audio and video inside Claude, you can use our MCP, $8.99 a month.

我也会在简介里放上这个服务的链接。

I'll leave a link in the description to that as well.

Meta刚刚在他们的AI眼镜上加了一个防篡改的LED灯，一旦光线被改动或破坏，摄像头就会自动关闭。

Meta just added a tamper proof LED to their AI glasses, and it disables the camera if the light is modified or destroyed.

这样你就知道，对方是不是在用Meta Ray-Ban或者其他Meta AI智能眼镜偷偷录像了。

So the idea being, you know, wanna know when someone's recording you with the Meta Ray Bans or any of these other Meta AI smart glasses.

很多人一直担心被人未经同意就拍下画面。

A lot of people have been complaining about, you know, non consensual recording.

人们在他们不知情或未同意的情况下偷偷录像。

People are recording them when they don't know or they don't consent.

有人在健身房里随便拍来拍去。

People are in the gym recording whatever.

所以他们加了一个指示灯，我觉得很多设备都有这种功能——一旦开始录制，灯就会亮，让你知道有人在拍。

And so, they've added a light, which I think a lot of devices have, like a light that pops on when it's recording, so you know someone's recording.

但理论上，有人拿根针之类的东西，把LED灯泡戳坏或者弄坏，对吧？

But theoretically, someone could go with like a pin and just like, you know, kind of shatter the LED bulb, or wreck it, right?

这样灯就不亮了，或者用东西遮住，搞点小动作。

So it doesn't turn on, or cover it, or do something.

所以现在他们做了防篡改设计，根本没法动手脚。

So they're making it so that you can't tamper with it.

只要灯坏了，设备就根本不会录像。

If it's broken, it won't actually record.

我觉得很多人都对这个设计特别满意。

And this is, I mean, I think a lot of people are really excited about that.

但就在同一周，Meta还在其他地方扩大数据收集范围。

But this is also happening the same week that Meta is expanding the data collection elsewhere.

他们默认允许AI训练使用公开的Instagram照片，还在测试一种眼镜，能持续采集音频，每几秒就拍一张照片——拜托，这太吓人了。

So they're gonna be allowing their AI to train on public Instagram photos by default, and they're testing glasses that continuously collect audio and they snap photos every few seconds, which, come on, that's terrifying.

不过也许他们是寄出了某种产品，觉得总有人会喜欢吧。

But I mean, maybe it's like they're mailed in a product that they think some people will like that.

我真不这么认为，但眼下这个正在测试的原型，正要把眼镜从按需拍摄的设备，

I really doubt that, but this particular prototype that is under testing is gonna move the glasses from being an on demand camera.

对吧？

Right?

变成一种始终开启的设备，内置持续工作的传感器，把不间断的音频和定期拍摄的照片直接传给Meta的AI系统。

So you're like, hey, Meta, record whatever I'm doing, it turns on, to actually just being always on and having an always on sensor, and it's feeding continuous audio and periodic photos to Meta's AI systems.

他们还没公布这款产品的上市时间。

They don't have a date launch for this.

我希望至少能有办法选择退出，不然我真不觉得这种眼镜还能继续受欢迎。

I don't I hope there'll be a way to opt out, otherwise I don't imagine these glasses will be continue to be very popular.

关于Instagram这件事，这个改动会影响大约十亿用户，也就是数十亿张照片。

For the Instagram thing, this is a change that'll affect about a billion, that, you know, billions of photos for users.

大多数发帖的人根本不想自己的照片被用来训练AI，这简直有点离谱。

Most like, everyone posting is not wanting their their photos to be trained, like AI to be trained on them, so that's gonna be kind of crazy.

OpenAI的首席未来学家约书亚·阿希姆，在公司待了九年之后离职了。

OpenAI's chief futurist, Joshua Achim, is leaving after about nine years.

他是公司最早期就加入的安全倡导者之一，资历最深。

He's one of the safety advocates that's been at the company the longest since the early days.

他离职了，这也是过去两年里第五位离开OpenAI的资深安全负责人。

He is leaving and he's the fifth senior safety focused leader to leave OpenAI in the last two years.

对吧？

Right?

他们正准备上市。

They're getting ready to go public.

他其实2017年是以实习生身份加入的，后来一直领导他们的使命对齐团队，直到这个团队被取消。

He actually joined as an intern back in 2017, and then he led their mission alignment team until it was disabled, or I guess, like, disbanded.

萨姆·阿尔特曼在今年二月就把它撤掉了。

Sam Altman got rid of it back in February of this year.

之后他转任首席未来学家，主要研究人工智能的利与弊。

Then he moved into chief futurist role and was kind of studying AI harms and benefits.

自2024年以来，已经有五位与安全相关的高层离开OpenAI，其中不少人还创办了专注于AI安全标准的非营利组织。

Five different safety adjacent leaders have now left OpenAI since 2024, and a lot of them have started non profits focused on AI safety standards.

前白宫AI顾问迪恩·鲍尔本周刚出任战略未来部门负责人，他将短暂与伊奇姆重叠工作，之后接替大量政策研究事务，但首席未来学家这个职位并不会另设人选。

The former White House AI advisor, Dean Ball, started this week as the head of strategic futures, and is gonna briefly overlap with Itchyme before he actually leaves, and will kind of take over a lot of policy focused research work without having a direct replacement for the chief futurist title.

如果你觉得自己特别适合这个职位，不妨给OpenAI投份简历。

So if you believe you'd be perfect for that job, maybe you should send in your application to OpenAI.

各位，非常感谢今天收听这期播客。

Guys, thank you so much for tuning into the podcast today.

别忘了去访问 aibox.ai。

Make sure to check out aibox.ai.

链接在简介里。

Link's in the description.

如果你还没给这个节目打分评价，那真的对我意义重大。

If you have not already left a review for the show, it would mean the world to me.

这能极大地帮助节目发展。

It helps the show out tremendously.

今年我计划了很多精彩的访谈和内容，都很值得期待。

I have some exciting interviews and content and stuff planned for the show this year.

如果你能留下评价，我会更有动力为你带来更棒的内容。

And if you would be able to leave a review, it would help me to bring you more incredible content.

在Spotify上，你得点开‘关于’这个标签。

Over on Spotify, you have to hit the About tab.

只要你听了三集以上，就可以评价了。

If you've listened to three episodes, you're eligible to leave a review.

没达到这个条件，系统是不让评的。

Otherwise, it won't let you.

如果你已经听了，那就太好了。

So if you have, that would be incredible.

点击关于页面。

Hit the About tab.

给几颗星。

Hit some stars.

在苹果上有点难找，你得去播客页面，然后就能点星星并留下评论。

On Apple, it's kind of annoying to find, but you go to the the podcast page and you can hit the stars and leave a comment.

我都会看。

I read them all.

我很感激。

I appreciate them.

好了，各位。

Alright, guys.

希望你们今天剩下的时间都过得特别棒。

I hope you all have a fantastic rest of your day.