Addressing ‘insider’ AI agents with Camille Stewart Gloster | The TechTank Podcast
TechTank, a biweekly podcast from the Center for Technology Innovation at Brookings, explores today’s most consequential technology issues.
TechTank 是布鲁金斯学会技术创新中心的双周播客，探讨当今最重要的技术问题。
Moderators Nicol Turner Lee and Darrell West speak with experts and policymakers to share data, ideas, and policy solutions that address the challenges of our digital world.
主持人 Nicol Turner Lee 和 Darrell West 与专家和政策制定者交谈，分享应对数字世界挑战的数据、想法和政策解决方案。
In the search for greater efficiency, organizations are increasingly turning to agentic AI.
为了寻求更高的效率，组织越来越多地转向代理人工智能。
These systems operate somewhat autonomously and can execute decisions and actions on behalf of humans, including scheduling meetings, making purchases, and generating various outputs.
这些系统在某种程度上自主运行，可以代表人类执行决策和行动，包括安排会议、进行购买和生成各种输出。
These more advanced capabilities also present more complex governance challenges.
这些更先进的功能也带来了更复杂的治理挑战。
Many of the issues have not been fully addressed as executives push the adoption of such cutting-edge technologies.
随着高管们推动采用此类尖端技术，许多问题尚未得到充分解决。
In the absence of federal guardrails on AI, it is critical for organizations and individuals to understand the technology, its limitations, and how it’s controlled.
在缺乏联邦对人工智能的监管的情况下，组织和个人了解这项技术、它的局限性以及它的控制方式至关重要。
In this episode of the TechTank podcast, co-host Nicol Turner Lee is joined by Camille Stewart Gloster, the CEO of CAS Strategies and former deputy national cyber director for technology and ecosystem security at the White House, to discuss the proposals for governing agents detailed in her new book, “ The Insider You Built: How Organizations Stay in Control of Autonomous AI Agents.
在 TechTank 播客的这一集中，联合主持人 Nicol Turner Lee 与 CAS Strategies 首席执行官、白宫前技术和生态系统安全国家网络副主任 Camille Stewart Gloster 一起讨论了她的新书《你构建的内部人士：组织如何保持对自主 AI 代理的控制》中详细介绍的治理代理提案。
“
“
Listen to the episode and subscribe to the TechTank podcast on Apple , Spotify , or Acast .
收听该剧集并订阅 Apple、Spotify 或 Acast 上的 TechTank 播客。
[00:00:00] CO-HOST NICOL TURNER LEE: You’re listening to TechTank, a biweekly podcast from the Brookings Institution exploring the most consequential technology issues of our time.
[00:00:00] 联合主持人 NICOL TURNER LEE：您正在收听 TechTank，这是布鲁金斯学会的双周播客，探讨我们这个时代最重要的技术问题。
From racial bias in algorithms to the future of work, TechTank takes big ideas and makes them accessible.
从算法中的种族偏见到未来的工作，TechTank 吸收了伟大的想法并让它们变得容易理解。
Welcome to the TechTank podcast.
欢迎来到 TechTank 播客。
I am co-host Nicol Turner Lee, senior fellow and director of the Center for Technology Innovation at the Brookings Institution.
我是布鲁金斯学会技术创新中心高级研究员兼主任尼科尔·特纳·李（Nicol Turner Lee）的共同主持人。
Wow, what a couple of weeks it’s been.
哇，已经过去几周了。
After the explosion of generative AI and some of the news headlines as what’s happening with regards to AI safety, this conversation I’m having today cannot be more timely.
在生成式人工智能爆炸式增长以及人工智能安全方面发生的一些新闻头条之后，我今天进行的这次对话再及时不过了。
[00:00:47] Companies and organizations are quickly implementing various tools around how they’re going to ensure that their AI agents don’t go rogue.
[00:00:47] 公司和组织正在快速实施各种工具，以确保他们的人工智能代理不会失控。
And now, as these agentic AI systems operate somewhat autonomously, this becomes a more ubiquitous concern, and enterprises are turning to a new set of tools to introduce different and more complex solutions to these governance challenges.
现在，随着这些代理人工智能系统在某种程度上自主运行，这成为一个更加普遍的问题，企业正在转向一套新的工具来引入不同的、更复杂的解决方案来应对这些治理挑战。
[00:01:12] For those of you who may not know, agentic systems can execute decisions and actions on behalf of humans.
[00:01:12] 对于那些可能不知道的人来说，代理系统可以代表人类执行决策和行动。
Yes, I said it.
是的，我说过。
Now listen, I’m not saying anything you don’talready know.
现在听着，我不会说任何你还不知道的事情。
This could include scheduling meetings, making appointments, purchases, generating other outputs.
这可能包括安排会议、预约、采购、产生其他产出。
The capabilities promise significant advances in efficiency.
这些功能有望显着提高效率。
[00:01:34] In fact, my doctor uses agentic AI for me to actually get an appointment.
[00:01:34] 事实上，我的医生使用代理人工智能为我实际预约。
But this does not come without its own risk, and that’s what we’re gonna talk about today.
但这并非没有风险，这就是我们今天要讨论的内容。
I’mjoined by my friend Camille Stewart Gloster, who is a researcher, writer, educator who works at the intersection of artificial intelligence, cybersecurity, and geopolitical risk in both the private and public sector.
我的朋友卡米尔·斯图尔特·格洛斯特 (Camille Stewart Gloster) 加入了我的行列，她是一位研究员、作家、教育家，在私营和公共部门的人工智能、网络安全和地缘政治风险领域工作。
[00:02:01] And I’m so proud to know her because she was the first deputy national cyber director for technology and ecosystem security at the White House, and the former global head of product security strategy at Google.
[00:02:01] 我很自豪认识她，因为她是白宫负责技术和生态系统安全的第一任国家网络副主任，也是谷歌前产品安全战略全球主管。
This range of experience informs her new book, “The Insider You Built: How Organizations Stay in Control of Autonomous AI Agents,” which I think, after looking at it, offers new answers to many of the critical questions that we’re asking today.
她的新书《你构建的内部人士：组织如何保持对自主人工智能代理的控制》中的这些经验为她提供了信息，我认为，在阅读完这本书之后，它为我们今天提出的许多关键问题提供了新的答案。
Camille, thank you for joining me.
卡米尔，谢谢你加入我。
[00:02:32] GUEST CAMILLE STEWART GLOSTER: Oh, it’s my pleasure.
[00:02:32] 嘉宾卡米尔·斯图尔特·格洛斯特：哦，这是我的荣幸。
Thank you for having me.
谢谢你有我。
I’m excited.
我很兴奋。
[00:02:35] CO-HOST NICOL TURNER LEE: I mean, listen.
[00:02:35] 联合主持人尼科尔·特纳·李：我的意思是，听着。
I know.
我知道。
Me too.
我也是。
I mean, your book was so timely, I have to say.
我的意思是，我不得不说，你的书来得太及时了。
[00:02:39] GUEST CAMILLE STEWART GLOSTER: You know, I wanted it to come out six months before it did, but I couldn’t have asked for a better moment where people would be more receptive to it than the one when it came out.
[00:02:39] 嘉宾卡米尔·斯图尔特·格洛斯特：你知道，我希望它能提前六个月问世，但我找不到比它问世时更容易接受的时刻了。
[00:02:49] CO-HOST NICOL TURNER LEE: And your book does something very unique.
[00:02:49] 联合主持人尼科尔·特纳·李：你的书做了一些非常独特的事情。
It sort of makes AI and cybersecurity bedfellows.
这在某种程度上让人工智能和网络安全成为了同床异梦的伙伴。
So I wanna start the podcast sort of diving into why did you write this book based on these experiences you’ve had, and tell us a little bit more about what’s in it.
因此，我想在播客开始时深入探讨您为何根据您的这些经历来写这本书，并告诉我们更多有关其中内容的信息。
[00:03:04] GUEST CAMILLE STEWART GLOSTER: Yeah.
[00:03:04] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
So when I left the White House, where we had just released the AI executive order and I was leading AI security, I started, helping companies and organizations of all sizes navigate AI deployment.
因此，当我离开白宫时，我们刚刚发布了人工智能行政命令，我负责领导人工智能安全，我开始帮助各种规模的公司和组织进行人工智能部署。
And what I quickly realized was they had no idea what they were doing.
我很快意识到他们不知道自己在做什么。
They were chasing innovation Andkind of sacrificing the infrastructure that we usually have when we either deploy a new capability or even more so as we move to agentic, as we allow another actor into our environment.
他们追求创新，并且牺牲了我们在部署新功能时通常拥有的基础设施，或者在我们转向代理时更是如此，因为我们允许另一个参与者进入我们的环境。
And while agents aren’t people, they can act.
虽然特工不是人，但他们可以采取行动。
They, they essentially make decisions.
他们，他们基本上做出决定。
They have permissions and an identity, at least they should, and the way they move through your environment is much like a person.
他们拥有权限和身份，至少他们应该如此，而且他们在您的环境中移动的方式很像一个人。
And this insider risk analogy became very prominent and one that I felt needed to be articulated more broadly for organizations to understand how to actually control this entity in your organization.
这种内部风险类比变得非常突出，我认为需要更广泛地阐明这一类比，以便组织了解如何实际控制组织中的该实体。
[00:04:09] We often have insider risk programs in companies that, um, kind of constrain the behavior of your staff and kind of direct it.
[00:04:09] 我们经常在公司中制定内部风险计划，嗯，这些计划会限制员工的行为并对其进行指导。
People make, uh, choices intentionally or unintentionally that can cause harm in an organization and the goal is to find that out and prevent it.
人们有意或无意地做出可能对组织造成伤害的选择，我们的目标是找出并防止它发生。
And we need to do the same thing for agents.
我们需要为代理商做同样的事情。
They are misconfigured or have emergent behavior or just have this by any means necessary mentality that means that what you th- the path you think they might take is not necessarily the path they’re gonnatake.
他们配置错误，或者有紧急行为，或者只是有这种无论如何必要的心态，这意味着你认为他们可能采取的路径不一定是他们将要采取的路径。
[00:04:42] CO-HOST NICOL TURNER LEE: Well, let’s stop for a minute because I think you and I know agents, like what they are, what they mean.
[00:04:42] 联合主持人尼科尔·特纳·李：好吧，让我们暂停一下，因为我想你和我都了解特工，知道他们是什么，他们的意思是什么。
[00:04:47] You know, can I tell you, whenever I think of an agent in AI, I think of these like minions with trench coats …
[00:04:47] 你知道，我可以告诉你，每当我想到人工智能中的代理时，我就会想到这些就像穿着风衣的小黄人......
kind of walking around.
有点四处走动。
You know what I mean?
你知道我的意思？
As, I cannot get that out of my head.
因为，我无法把它从我的脑海中抹去。
I think somebody said this a couple of years ago and all I see are like these little things with trench coats, like without no human supervision.
我想几年前有人说过这句话，我看到的只是这些穿着风衣的小东西，就像没有人类监督一样。
Tell folks what do you mean in the book by agents?
告诉大家你在书中所说的代理人是什么意思？
So, because there’s so many definitions, Camille, about, you know, agentic infrastructure that I find it to be quite interesting that it’s gonna get bigger and bigger …
所以，卡米尔，关于代理基础设施有很多定义，我发现它非常有趣，它会变得越来越大……
in terms of functions, you know?
就功能而言，你知道吗？
[00:05:16] GUEST CAMILLE STEWART GLOSTER: I’m so glad you asked that.
[00:05:16] 嘉宾卡米尔·斯图尔特·格洛斯特：我很高兴你问这个问题。
I liked your minions.
我喜欢你的小黄人。
[00:05:19] CO-HOST NICOL TURNER LEE: Yes, it’s, it’s a minions.
[00:05:19] 联合主持人尼科尔·特纳·李：是的，它是，它是一个小黄人。
[00:05:21] I know I said it first, but I, I’m not the owner of that analogy, okay?
[00:05:21] 我知道我先这么说，但我，我不是这个比喻的主人，好吗？
Okay.
好的。
Somebody else said it.
别人也这么说过。
[00:05:25] GUEST CAMILLE STEWART GLOSTER: Okay, fair.
[00:05:25] 嘉宾卡米尔·斯图尔特·格洛斯特： 好的，公平。
Fair.
公平的。
Um, you know, one of the things I talk about in the book is that I don’t want people to quibble over the definition.
嗯，你知道，我在书中谈到的一件事是我不希望人们对定义争论不休。
To your point, there are a wide range of definitions of what is agentic, and I encourage people to focus on the point at which The tool is taking action without immediate direction.
就您的观点而言，对于什么是代理有很多种定义，我鼓励人们关注该工具在没有立即指示的情况下采取行动的点。
So not an if-then statement.
所以不是 if-then 语句。
You know, we all remember being able to set up, like, if this happens, then that happens.
你知道，我们都记得能够设置，就像，如果发生这种情况，那么就会发生这种情况。
Not that, but where a tool is actually synthesizing information, analyzing it, and can make a bevy of choices to get to an outcome.
不是这样，而是工具实际上正在综合信息、分析信息，并且可以做出一系列选择以获得结果。
When you have that, I don’t care if it’s got the harness, if it doesn’t have the harness, if it…
当你拥有它时，我不在乎它是否有安全带，是否没有安全带，是否......
[00:06:06] You are in an agentic territory and should be thinking about how you govern that behavior and understand that behavior and have visibility into that behavior as if it is the most sophisticated of agents.
[00:06:06] 您处于代理领域，应该考虑如何管理该行为、理解该行为并了解该行为，就好像它是最复杂的代理一样。
[00:06:19] CO-HOST NICOL TURNER LEE: Well, and it’s so interesting to me.
[00:06:19] 联合主持人尼科尔·特纳·李：嗯，这对我来说很有趣。
I mean, there are high numbers of industry right now that are deploying agents, right?
我的意思是，现在有很多行业正在部署代理，对吗？
Like I mentioned, whether it’s the appointment agent, right, or it’s an agent that is helping to sort of synthesize a variety of data inputs for the, uh, company, you know, wants to make sure they have better client efficiency.
就像我提到的，无论是预约代理，对吧，还是帮助合成各种数据输入的代理，呃，公司，你知道，希望确保他们有更好的客户效率。
I find it so interesting that companies have sort of moved quickly on the agentic side to some respects.
我觉得很有趣的是，公司在某些方面在代理方面采取了快速行动。
Not all, but to some respect.
不是全部，但在某些方面。
So the question I have for you is, like, work- are companies, like, prepared for this?
所以我要问你的问题是，比如工作——公司是否为此做好了准备？
I mean, I thought we were still just trying to figure out predictive and generative AI and now- …
我的意思是，我以为我们仍在试图找出预测性和生成性人工智能，而现在——……
you know, it’s sort of like they’ve leapfrogged to the use of agents.
你知道，这有点像他们已经跨越到使用代理。
I mean, where are companies when it comes to the use of these technologies?
我的意思是，在使用这些技术方面，公司处于什么位置？
Because I think your book is very relevant around, you know, managing security risk as well.
因为我认为你的书与管理安全风险也非常相关。
[00:07:08] GUEST CAMILLE STEWART GLOSTER: Yeah, I mean, one of the things the book talks about is kind of breaking down these silos between security, safety, enterprise risk management, all of those things because the largest instance of shadow AI right now is increasingly agentic.
[00:07:08] 嘉宾卡米尔·斯图尔特·格洛斯特（CAMILLE STEWART GLOSTER）：是的，我的意思是，这本书谈到的一件事是打破安全、安全、企业风险管理之间的孤岛，所有这些事情都是因为目前最大的影子人工智能实例越来越代理。
[00:07:23] GUEST CAMILLE STEWART GLOSTER: You know, many of these generative AI capabilities, you give your, your team Claude and it’s got Claude Cowork, ChatGPT Work, all of these capabilities that can look across systems, pull data, analyze that data, make decisions, take actions on your behalf.
[00:07:23] 嘉宾卡米尔·斯图尔特·格洛斯特（CAMILLE STEWART GLOSTER）：你知道，你给你的团队 Claude 提供了许多生成式 AI 功能，它有 Claude Cowork、ChatGPT Work，所有这些功能都可以跨系统查看、提取数据、分析数据、做出决策、代表你采取行动。
Most organizations were kind of chasing or following the innovation, trying to experiment and didn’t wrap their, take the time to wrap their heads around the, the full scope of the problem at first.
大多数组织都在追逐或追随创新，尝试进行实验，但一开始并没有花时间思考问题的全部范围。
And there are organizations that are really leaning in to continue to learn and be agile around, okay, well what does this new discovery mean?
有些组织确实倾向于继续学习并保持敏捷，好吧，这个新发现意味着什么？
What does this new action or behavior mean for how we govern and understand agents in our environment?
这种新的行动或行为对于我们如何管理和理解环境中的主体意味着什么？
And some are just kind of hoping for the best, which is the scariest of territories.
有些人只是希望有最好的结果，这是最可怕的领域。
[00:08:11] CO-HOST NICOL TURNER LEE: Well, and what I like about your book is, I mean, you’re a policy woman as well, right?
[00:08:11] 联合主持人尼科尔·特纳·李：嗯，我喜欢你的书的一点是，我的意思是，你也是一名政策女性，对吧？
I mean, this is not just, you know, Camille as the cyber expert.
我的意思是，这不仅仅是卡米尔作为网络专家。
You’ve had a lot of experience thinking about the policies that move any type of technology.
您在思考推动任何类型技术发展的政策方面拥有丰富的经验。
‘Cause I’m sure when you were in the Obama administration, you weren’ttalking about agentic AI.
因为我确信当你在奥巴马政府时，你并没有谈论代理人工智能。
[00:08:28] Uh, okay, right, I was gonna say, I think we were just talking about cyber- We weren’t even there yet.
[00:08:28] 呃，好吧，对，我想说，我想我们只是在谈论网络——我们还没到那一步。
We weren’t even there, right?
我们根本就不在那儿，对吧？
I mean, I was, like, thinking about it the other day when I knew I was interviewing you.
我的意思是，有一天，当我知道我正在采访你时，我就在想这个问题。
So let’s think about it for just a second.
让我们考虑一下。
I mean, most of the framework that’s been out there when it comes to technology, um, the, well, let’s just say the safety and security of technology innovation, has been within a cybersecurity framework or resilience framework.
我的意思是，大多数涉及技术的框架，嗯，好吧，我们只是说技术创新的安全性和保障，一直在网络安全框架或弹性框架内。
I’d love to hear from you before I go a little bit more into the details for those who are listening around, you know, what you’re suggesting they should do, uh, to make sure that they’re, uh, managing the risk of agentic AI.
在我为那些正在倾听的人详细介绍之前，我很想听听您的意见，您知道，您建议他们应该做什么，呃，以确保他们，呃，管理代理人工智能的风险。
But, like, why is it so important, Camille, to have this conversation around what cyber looks like today in the age of AI?
但是，卡米尔，为什么围绕人工智能时代今天的网络是什么样子进行对话如此重要？
‘Cause I think the book really brings that out, too, in terms of governance is really required as well.
因为我认为这本书也确实阐述了这一点，在治理方面也确实需要。
[00:09:14] GUEST CAMILLE STEWART GLOSTER: Yes.
[00:09:14] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
So first, I decided to focus on organizations because we are in a policy moment where policy is not moving fast enough.
因此，首先，我决定关注组织，因为我们正处于一个政策进展不够快的政策时刻。
Um, we’re not wrapping our heads around it enough.
嗯，我们对这个问题的理解还不够。
And in certain places, like here, we’re paused on action at a national level in terms of a concrete AI policy.
在某些地方，比如这里，我们在国家层面暂停了具体人工智能政策的行动。
And that means that industry is on the front lines of writing the rules and the societal norms and the way this technology is going to impact our lives.
这意味着行业处于制定规则和社会规范以及这项技术将如何影响我们生活的最前沿。
And unwittingly, organizations have stepped into a role that they weren’t designed to fill.
不知不觉中，组织已经扮演了一个他们原本不应该扮演的角色。
But also, even if you think about this from a maximizing business value, getting the m- the best return on investment, these decisions that they need to make, these investments in governance, and I, I’m trying to expand the term governance ’cause when you hear governance, you think compliance, risk, policy only.
而且，即使你从最大化商业价值、获得最佳投资回报、他们需要做出的这些决策、这些对治理的投资来考虑这个问题，我也在尝试扩展治理这个术语，因为当你听到治理时，你只会想到合规性、风险和政策。
But when I say governance, I truly mean, like, wrapping your hands around this capability.
但当我说治理时，我真正的意思是，将你的双手掌握在这种能力上。
And it…
而它……
That’s why it’s such a cross-functional discipline.
这就是为什么它是一个跨职能学科。
You need security tooling, security frameworks, um, security mindset, but you also need to understand enterprise risk management, all of the safety features, um, the ethics questions.
您需要安全工具、安全框架，嗯，安全思维，但您还需要了解企业风险管理，所有安全功能，嗯，道德问题。
[00:10:32] All of those things come to play in an agentic environment because it is very reminiscent of an organizational environment writ large.
[00:10:32] 所有这些事情都在代理环境中发挥作用，因为它很容易让人想起组织环境。
And you’ve got all of these disciplines, right?
你已经掌握了所有这些学科，对吗？
You- when you hired people, you’ve got HR, you’ve got legal, you’ve got recruiting that brought people in.
当你雇佣员工时，你有人力资源部门，你有法律部门，你有招聘人员来吸引人才。
You’ve got a manager to oversee them.
你有一位经理来监督他们。
How do we create that infrastructure for this new actor in our environment?
我们如何在我们的环境中为这个新参与者创建基础设施？
And cybersecurity is at the center of that, yes, but it does not act alone.
是的，网络安全是其中的核心，但它并不是单独行动的。
[00:11:02] CO-HOST NICOL TURNER LEE: Yeah, I mean, it makes sense.
[00:11:02] 联合主持人尼科尔·特纳·李：是的，我的意思是，这是有道理的。
I mean, at the governance level, is this a conversation that policymakers need to be having now, right?
我的意思是，在治理层面，这是政策制定者现在需要进行的对话，对吧？
Because it appears with all of the things that have been going on lately in the news, uh, like, you know, OpenAI agents recently, I think, hacked into a UN, um, website and- Yes Australia was actually a victim.
因为它与最近新闻中发生的所有事情一起出现，呃，就像，你知道，OpenAI 特工最近，我认为，入侵了联合国，嗯，网站，并且 - 是的，澳大利亚实际上是受害者。
I mean, how do policymakers sort of reconcile what security now looks like, you know, both from a technical standpoint but also a policy standpoint?
我的意思是，政策制定者如何从技术角度和政策角度协调现在的安全状况？
[00:11:29] GUEST CAMILLE STEWART GLOSTER: We have to get better situated to have more ali- agile policy development.
[00:11:29] 嘉宾卡米尔·斯图尔特·格洛斯特：我们必须更好地制定更加灵活的政策。
We love to get it right, and we are letting perfection be the enemy of good.
我们喜欢把事情做好，但我们却让完美成为优秀的敌人。
And one of the things that I think we’ve started to see in how the EU does policymaking is they write something, and then they know you’re gonna react to it,everyone’s gonna react to it.
我认为我们已经开始看到欧盟如何制定政策的一件事是他们写了一些东西，然后他们知道你会对此做出反应，每个人都会对其做出反应。
And they make tweaks to it and adjust.
他们会对其进行调整和调整。
And That adjustment needs to continue even after the legislation has gone into effect.
即使立法生效后，这种调整也需要继续。
And that is a model we need to start to adopt more broadly, that we take some time to understand and collaborate with industry, collaborate with academia to understand the long-term impacts, this kind of cross-sector development of policy.
这是我们需要开始更广泛采用的模式，我们需要一些时间来了解并与行业合作，与学术界合作以了解这种跨部门政策制定的长期影响。
But we’ve got to develop something, and we definitely have to articulate the red lines, and that is work we have not yet done.
但我们必须开发一些东西，我们肯定必须阐明红线，而这是我们尚未完成的工作。
And so our organizations, our, our industry is kind of writing its own rules and actually does want that guidance.
因此，我们的组织、我们的行业正在制定自己的规则，并且实际上确实需要这种指导。
[00:12:34] CO-HOST NICOL TURNER LEE: Yes.
[00:12:34] 联合主持人尼科尔·特纳·李：是的。
Yes.
是的。
Yes.
是的。
But then it’s so interesting, too, if I can, um, you are a cybersecurity person, and I don’t know, the cybersecurity people have been around for a long time and have their own norms and standards, right?
但这也很有趣，如果我可以的话，嗯，你是一个网络安全人员，我不知道，网络安全人员已经存在很长时间了，有自己的规范和标准，对吧？
Yes.
是的。
And then in comes AI.
然后是人工智能。
So from the standpoint of that community, are there any changes or adaptations that they need to continue to make?
那么从该社区的角度来看，他们是否需要继续做出任何改变或适应？
Because I, I just, I just think that AI has not come with a set of clearly defined, um, norms, uh, which makes it so hard, right, to do policy.
因为我，我只是，我只是认为人工智能还没有一套明确定义的，嗯，规范，呃，这使得制定政策变得如此困难。
You know, I was just talking about this with somebody.
你知道，我刚刚和某人谈论过这件事。
It’s not like the internet, you know what I mean?
这不像互联网，你明白我的意思吗？
It’s something different.
这是不同的东西。
So I’m just curious from you, like, the cybersecurity community, who’s so much leaning into ISO language and other things- …
所以我只是对你们（比如网络安全社区）很好奇，他们非常倾向于 ISO 语言和其他东西——……
Camille, right?
卡米尔，对吗？
I mean, how are they, like, handling and managing this?
我的意思是，他们是如何处理和管理这个问题的？
And, like, on their terms, like, what would be ideal for them to be better stewards in this, uh, new, innovative period?
而且，按照他们的说法，在这个，呃，新的创新时期，对于他们来说，怎样才能成为更好的管理者？
[00:13:28] GUEST CAMILLE STEWART GLOSTER: Yeah.
[00:13:28] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
I mean, everybody’s gotta be more agile.
我的意思是，每个人都必须变得更加敏捷。
So yes, the discipline will change.
所以是的，纪律将会改变。
A lot of the foundations remain the same, though.
不过，许多基础保持不变。
One of the capabilities that organizations need to develop is platform engineering.
组织需要开发的能力之一是平台工程。
Engineering with AI agents is going to be an even bigger part of advancing security and safety because you’ve gotta be able to actually make the adjustments needed.
人工智能代理的工程将成为提高安全性的更重要部分，因为你必须能够实际进行所需的调整。
Governance requires runtime enforcement, which means enforcing your governance principles, your rules at the point of action.
治理需要运行时执行，这意味着在行动时执行您的治理原则和规则。
It cannot be that you just write a policy and it lives out in the ether.
不可能你只写了一份保单，它就在空中消失了。
You’ve gotta combine that technical apparatus to say, as an agent is making decisions, we’re able to flag that this one needs additional review, this one needs, can move forward, and this one we need to just shut down immediately.
你必须结合该技术设备来说，当代理做出决策时，我们能够标记这个需要额外的审查，这个需要，可以继续前进，而这个我们需要立即关闭。
And, um, and so those are some of the the changes that I, I would like to see, as well as this kind of moving more quickly.
嗯，这些是我希望看到的一些变化，以及这种更快的进展。
You know, in security we’ve got these 30, 60, 90-day patch cycles to patch software and to close security bugs.
您知道，在安全方面，我们有 30 天、60 天、90 天的补丁周期来修补软件并消除安全错误。
Those things take too long.
这些事情需要太长时间。
Continuous learning, this, this aspect of seeing an incident unfold, seeing an emergent behavior in an agent, uh, make itself available, seeing a misconfiguration and changing it across all of those different s- parts of the stack, organizational, operational, and technical, as quickly as possible is muscle memory that organizations don’t come to easily, but we have to adjust to.
持续学习，这个，看到事件展开的这个方面，看到代理中的紧急行为，呃，让自己变得可用，看到错误配置并在堆栈的所有不同部分（组织、操作和技术）中对其进行更改，尽快是组织不容易掌握的肌肉记忆，但我们必须适应。
[00:15:00] CO-HOST NICOL TURNER LEE: So you talk about this in the book, um, this proposed framework, the authority-centered enforcement and attribution, or ACE framework.
[00:15:00] 联合主持人 NICOL TURNER LEE：所以你在书中谈到了这个，嗯，这个提议的框架，以权威为中心的执行和归因，或 ACE 框架。
Explain to us what that is ’cause I think what you’re talking about and why I really like the book is that you’re giving people tools to be able to navigate this space of uncertainty, which I think is really helpful, right, at this point in time.
向我们解释一下这是什么，因为我认为你在谈论的内容以及我真正喜欢这本书的原因是，你为人们提供了能够驾驭这个不确定空间的工具，我认为这在此时此刻确实很有帮助。
[00:15:20] GUEST CAMILLE STEWART GLOSTER: Yes.
[00:15:20] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
ACE is that governance in action that I’m talking about.
ACE 就是我所说的行动治理。
[00:15:25] GUEST CAMILLE STEWART GLOSTER: There are seven steps, but they basically break down to articulating your authority.
[00:15:25] 嘉宾卡米尔·斯图尔特·格洛斯特：有七个步骤，但它们基本上可以分解为阐明你的权威。
So how much authority do you wannadelegate to this tool?
那么您想向这个工具授予多少权限？
And that doesn’t just mean I want it to do X.
这不仅仅意味着我希望它做某件事。
It also means I don’t want it to do these things.
这也意味着我不希望它做这些事情。
Here’s where it should stop.
这是它应该停止的地方。
Here’s how, um, much money it should be able to spend or, you know, how long it should have access to a system to be able to achieve the goal that I’ve given it, all of those things.
这是它应该能够花多少钱，或者，你知道，它应该能够访问系统多长时间才能实现我给它的目标，所有这些事情。
It is thinking through the detection and governance of it.
它正在通过对其的检测和治理进行思考。
So what are the broader system-wide aspects that we need?
那么我们需要哪些更广泛的全系统方面呢？
How do we understand its identity?
我们如何理解它的身份呢？
What permissions should it have?
应该有什么权限？
Um, who is managing it?
嗯，谁在管理？
We need to know who is accountable within the organization for that agent, for thinking through its threat model and communicating that to the security and the IT team so that continues to evolve as you understand its ability to have impact on your customers and on your infrastructure in turn.
我们需要知道组织内谁对该代理负责，负责思考其威胁模型并将其传达给安全和 IT 团队，以便随着您了解其对客户和基础设施产生影响的能力而不断发展。
And then it’s that runtime enforcement.
然后就是运行时强制执行。
How do we ensure that at the point of action we can pull back a behavior, that we can identify where, um, an agent’s actions are going to deviate with, from the objective that was set?
我们如何确保在行动时我们可以撤回某种行为，我们可以确定代理的行动将在哪里偏离既定的目标？
And then the last part is thatcontinuous learning.
最后一部分是持续学习。
How do we make sure we adapt as we gain new information?
当我们获得新信息时，我们如何确保我们能够适应？
[00:16:48] Whether it’s because you saw the OpenAI Hugging Face incident in the news and you learned some tools and tactics from that, or because within your own organization you discovered a misconfiguration or an emergent behavior or a mechanism to achieve the objective that does not align with how your organization wants to see it happen.
[00:16:48] 无论是因为您在新闻中看到了 OpenAI Hugging Face 事件并从中学到了一些工具和策略，还是因为您在自己的组织内发现了错误配置、紧急行为或实现目标的机制与您的组织希望看到的方式不一致。
And sothe ACE framework really makes those things move with the agent so that as it takes action, as it has authority within your organization, you’re seeing all of those things in real time.
因此，ACE 框架确实使这些事物与代理一起移动，以便当它采取行动时，因为它在您的组织内拥有权威，您可以实时看到所有这些事物。
[00:17:22] CO-HOST NICOL TURNER LEE: Well, and I think that’s such a useful tool.
[00:17:22] 联合主持人尼科尔·特纳·李：嗯，我认为这是一个非常有用的工具。
I mean, that this is for business people, right?
我的意思是，这是为商务人士准备的，对吗？
In many respects, you’re giving them agency.
在很多方面，你给了他们代理权。
You’re pulling them into the process.
你正在将他们拉入这个过程。
[00:17:30] GUEST CAMILLE STEWART GLOSTER: Yes.
[00:17:30] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
And so this book is not just for cybersecurity professionals.
因此，这本书不仅仅适合网络安全专业人士。
Right.
正确的。
Right.
正确的。
It’s for everyone in the organization.
它适用于组织中的每个人。
That marketing person who now has been charged to create a s- customer success bot and, um, is, has on the front lines of seeing how it interacts with the customers, what kind of data it has access to, but also recognizes just how important it is to be managing this tool.
现在负责创建客户成功机器人的营销人员，嗯，是在第一线了解它如何与客户交互，它可以访问什么样的数据，但也认识到管理这个工具是多么重要。
[00:17:51] This book will help you know your place in that broader ecosystem and even advocate to your leadership for more governance to build out the infrastructure that I’m talking about.
[00:17:51] 这本书将帮助您了解自己在更广泛的生态系统中的位置，甚至向您的领导层倡导更多的治理，以构建我正在谈论的基础设施。
There’s a chapter in the book about the business case ’cause what I’m finding is whether it’s security or a business unit frontline owner who owns a agent- They often are convincing their leadership that they need the resources, the restructuring, the access to information within their organization to actually be able to do these governance actions.
书中有一章是关于业务案例的，因为我发现无论是安全部门还是拥有代理的业务部门一线所有者——他们经常说服领导层，他们需要资源、重组、访问组织内的信息，以便真正能够执行这些治理行动。
[00:18:23] CO-HOST NICOL TURNER LEE: And then I would just think, um, you know, ’cause of your background working in this space at, in the public interest level, um, what about, um, or at the public interest level, what about people like my mother?
[00:18:23] 联合主持人尼科尔·特纳·李：然后我会想，嗯，你知道，因为你在这个领域工作的背景，在公共利益层面，嗯，或者在公共利益层面，像我母亲这样的人呢？
Yeah.
是的。
Are there things that they need to know?
有他们需要知道的事情吗？
I mean, I was just in a conversation, and I think peopleare throwing their hands up that this AI stuff is gonna be hard to do traditional, you know, consumer education.
我的意思是，我刚刚在一次谈话中，我认为人们举手表示，这种人工智能的东西很难进行传统的消费者教育。
But I do think, you know, with the amount of scams that are happening and all of the technology moving so rapidly, consumers also need to have some agency, much like the businesses.
但我确实认为，随着诈骗数量的不断发生以及所有技术的快速发展，消费者也需要有一些代理机构，就像企业一样。
So just curious from you, like, what can consumers do as well to manage this growing agentic infrastructure?
您很好奇，消费者还能做什么来管理这个不断增长的代理基础设施？
Or maybe even, like, let the companies that they, um, you know, are interacting with know that something happened with their agent.
或者甚至，比如，让与他们，嗯，你知道，正在互动的公司知道他们的代理人发生了一些事情。
[00:19:08] Yeah.
[00:19:08] 是的。
I don’t know.
我不知道。
I’m just, I’m kind of, like, brainstorming here, Camille, right?
我只是，我有点像在这里集思广益，卡米尔，对吧？
[00:19:10] GUEST CAMILLE STEWART GLOSTER: Yes.
[00:19:10] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
Well, so there’s a bonus chapter focused on individuals and a distillation of the framework for your everyday life.
嗯，所以有一个额外的章节重点关注个人以及日常生活框架的提炼。
Yes.
是的。
Sowhether you’re using an agent to book travel or using it to manage your life, you need to be thinking about four things.
因此，无论您是使用代理来预订旅行还是使用它来管理您的生活，您都需要考虑四件事。
[00:19:24] What permission are you giving it?
[00:19:24] 你给它什么许可？
Like, similar to what I talked about with authority.
就像，类似于我权威谈论的内容。
What do you want it to do?
你想让它做什么？
What don’t you want it to do?
你不希望它做什么？
What are the boundaries?
界限是什么？
Um, how is, how are you gonna be accountable for the actions it takes, and are you comfortable with that?
嗯，怎么样，你将如何对所采取的行动负责，你对此感到满意吗？
So if that travel agent, um, buys 15 tickets rather than one ’cause you wanted it to check for the next 15 days for the best price, and it took that as buy a ticket every day for the next 15 days rather than just buy the best price over the course of 15 days, um, are you f- comfortable being accountable to that?
因此，如果该旅行社购买 15 张机票而不是一张，因为您希望它检查接下来 15 天的最优惠价格，并且在接下来的 15 天内每天购买一张机票，而不是只在 15 天内购买最优惠价格，那么您是否愿意对此负责？
And if not, then how can you make sure that it only buys one ticket or it only spends X amount of dollars?
如果不是，那么你如何确保它只购买一张票或者只花费 X 美元？
What is the broader context in which it operates?
它运作的更广泛背景是什么？
Make sure you’re giving…
确保你给予……
The prompt has enough information, that you’ve given it enough context, that you understand the context enough to, to see what could be harmful.
该提示包含足够的信息，您已经为其提供了足够的上下文，您足以了解上下文，以了解哪些内容可能有害。
And then what is the evolution of the technology?
那么技术的演变是怎样的呢？
And one of the things people say is, “Oh my gosh, it’s moving so fast.
人们说的其中一件事是，“天哪，它发展得太快了。
How do I keep up and then be able to adjust my behaviors?” I mean, use the tool to help you understand the tool.
我如何跟上并调整我的行为？”我的意思是，使用该工具可以帮助您了解该工具。
Set up a prompt that tells you once a week, once a month, what are the new capabilities that I should be worried about or thinking about, or adjust my behavior relative to these tasks that I have running And then use that to guide how you’re thinking about it in addition to you consuming the news and understanding the capability.
设置一个提示，每周一次、每月一次，告诉你我应该担心或考虑的新功能是什么，或者根据我正在运行的这些任务调整我的行为，然后用它来指导你除了消费新闻和理解功能之外如何思考它。
[00:20:46] So yes, I thought very much about the individual.
[00:20:46] 所以，是的，我非常关心个人。
That’s why even in the organizational context, it’s written so that each person could use it to navigate.
这就是为什么即使在组织环境中，它也是这样编写的，以便每个人都可以使用它进行导航。
And then this bonus chapter with the PACE framework, the personal ACE framework, um, helps people think about this in context.
然后这个奖励章节涉及 PACE 框架，个人 ACE 框架，嗯，帮助人们在上下文中思考这个问题。
But you’re right, we have to change our mindset.
但你说得对，我们必须改变我们的心态。
We have to …
我们必须……
stop this.
停止这个。
[00:21:05] CO-HOST NICOL TURNER LEE: We do.
[00:21:05] 联合主持人尼科尔·特纳·李：我们愿意。
We do.
我们做到了。
And look, I need you to come to my house and teach me how to do some of this optimization stuff for the bot.
看，我需要你来我家教我如何为机器人做一些优化工作。
I haven’tfigured it out, Camille.
我还没弄清楚，卡米尔。
I mean, I’m, I, I do think though once I do, your framework is gonna be helpful because I think we do not want to see people, you know, have the agents go out of control and then the next thing you know, you know, we’re sitting here trying to figure out what to do next.
我的意思是，我，我，我确实认为，一旦我这样做，你的框架将会有所帮助，因为我认为我们不想看到人们，你知道，让特工失去控制，然后你知道的下一件事，你知道，我们坐在这里试图弄清楚下一步该做什么。
I mean, which, you know, it also brings me…
我的意思是，你知道，这也给我带来了……
I just read an article with Bill Gates, essentially said, you know, a kill switch is not enough.
我刚刚读了比尔盖茨的一篇文章，基本上是说，你知道，一个终止开关是不够的。
Yeah.
是的。
Because I think of all these things that you’re talking about.
因为我想到了你正在谈论的所有这些事情。
And I was thinking for a while, kill switch should be good.
我想了一会儿，终止开关应该不错。
Somebody should be able to have, you know, the phone back in the days where you are able to push the button.
你知道，在你能够按下按钮的日子里，应该有人能够拥有手机。
You know, somebody should push the button to say, “The agents are out, go- out of control,” right?
你知道，应该有人按下按钮说：“特工出去了，失控了，”对吗？
But what do you think about that?
但你对此有何看法？
Like, actually having a kill switch is not the answer.
就像，实际上拥有一个终止开关并不是答案。
It really goes back to what you’ve been talking about, right?
这确实又回到了你刚才所说的，对吧？
[00:21:56] GUEST CAMILLE STEWART GLOSTER: Yes.
[00:21:56] 嘉宾卡米尔·斯图尔特·格洛斯特： 是的。
We need so much more than just a kill switch.
我们需要的不仅仅是一个终止开关。
I mean, uh- Realistically, you might be able to have a kill switch within your organization, but what would a kill switch look like across society?
我的意思是，呃-实际上，你也许可以在你的组织内设置一个终止开关，但是整个社会的终止开关会是什么样子呢？
I mean, it would- Yeah …
我的意思是，它会——是的……
essentially mean shutting down the internet, right?
本质上意味着关闭互联网，对吗？
Yeah.
是的。
You’d have to get it, no access to compute, no access to the internet and other data.
你必须得到它，无法进行计算，无法访问互联网和其他数据。
Like, that’s a really tough proposition.
就像，这是一个非常艰难的提议。
So yes, you should have kill switches within your organization and be able to shut down an agent.
所以，是的，您的组织内应该有终止开关，并且能够关闭代理。
You as an individual should know how to shut down your agents.
作为个人，您应该知道如何关闭您的代理。
But we need far more than that.
但我们需要的远不止于此。
You need to be consistently training it to understand your objectives and your thinking, that kind of compounding learning of what you want it to do and where you want its boundaries to be.
你需要不断地训练它来理解你的目标和你的想法，这种复合学习你想要它做什么以及你想要它的边界在哪里。
You need to be continuing to learn how to prompt better and how to, you know, provide the necessary context and understand how it’s evolving, and really keep pressure testing what are the boundaries of the permission I gave it.
你需要继续学习如何更好地提示，以及如何，你知道，提供必要的背景并了解它是如何演变的，并真正保持压力测试我所授予的权限的界限。
[00:22:50] CO-HOST NICOL TURNER LEE: Love it.
[00:22:50] 联合主持人尼科尔·特纳·李：喜欢它。
You know, I have to ask this question as we wrap up.
你知道，在我们结束时我必须问这个问题。
What was your experience writing your book?
您写书的经历是什么？
I always ask that question.
我总是问这个问题。
Uh, you know I’m an author, too.
呃，你知道我也是一名作家。
Yeah.
是的。
I felt quite lonely at some points, and so now I ask this question when we have authors that come on.
在某些时候我感到很孤独，所以现在当我们有作家出现时我会问这个问题。
[00:23:05] CO-HOST NICOL TURNER LEE: What was it like to write your book?
[00:23:05] 联合主持人尼科尔·特纳·李：写书的感觉如何？
[00:23:08] GUEST CAMILLE STEWART GLOSTER: Lonely and urgent.
[00:23:08] 嘉宾卡米尔·斯图尔特·格洛斯特：孤独而紧迫。
I mean, it, I was balancing everyday work, parenthood, family, all these things, with writing this book, and a recognition that the technology is moving so fast that this book could be obsolete by the time it came out.
我的意思是，我在写这本书的同时平衡了日常工作、为人父母、家庭以及所有这些事情，并认识到技术发展如此之快，以至于这本书在出版时可能已经过时了。
[00:23:25] GUEST CAMILLE STEWART GLOSTER: And the publishing process, as you know, is so long.
[00:23:25] 嘉宾卡米尔·斯图尔特·格洛斯特： 如您所知，发布过程非常漫长。
It can be two years, depending on the publisher.
可能是两年，具体取决于出版商。
So, um, I felt like I w- itwas a race against time.
所以，嗯，我觉得这是一场与时间的赛跑。
I was dedicating a lot of time, and I’m just glad it hit at a moment where people were ready to receive it.
我投入了很多时间，我很高兴它出现在人们准备好接受它的时刻。
[00:23:41] CO-HOST NICOL TURNER LEE: The timing is just impeccable with regards to the knowledge that you’re bringing to both enterprises, individuals, as well as, uh, society writ large.
[00:23:41] 联合主持人尼科尔·特纳·李：就您为企业、个人以及整个社会带来的知识而言，时机是完美的。
I mean, we have a lot of work to do in this area …
我的意思是，我们在这个领域还有很多工作要做……
Camille.
卡米尔。
And so I really appreciated the book and I appreciate you for being that voice that is able to bring these communities together, because it’s not gonna get, um, what do they say?
所以我真的很欣赏这本书，也很感谢你能够将这些社区聚集在一起，因为它不会得到，嗯，他们说什么？
It’s gonna get worse before it gets better.
在好转之前，情况会变得更糟。
Yeah.
是的。
So we do need to be paying attention to this.
所以我们确实需要关注这一点。
Um, thank you so much for joining me today and discussing your recent book.
嗯，非常感谢您今天加入我并讨论您最近的书。
Um, I ex- anticipate so much more from you.
嗯，我对你期待更多。
But where can people get “The Insider You Build”?
但人们可以从哪里获得“The Insider You Build”呢？
[00:24:15] GUEST CAMILLE STEWART GLOSTER: It’s available everywhere, but if you go to insideryoubuilt.com, you can find out everything about the book.
[00:24:15] 嘉宾卡米尔·斯图尔特·格洛斯特：随处可见，但如果您访问 insideryoubuilt.com，您可以找到有关这本书的所有信息。
[00:24:21] CO-HOST NICOL TURNER LEE: Perfect.
[00:24:21] 联合主持人尼科尔·特纳·李：完美。
And I know you’ve been out there.
我知道你去过那里。
Uh, they can find out where you’re gonna be doing book talks and all that other stuff.
呃，他们可以知道你要在哪里进行读书讲座以及所有其他事情。
Yes.
是的。
Uh, where do they find that out?
呃，他们从哪里发现的？
[00:24:28] GUEST CAMILLE STEWART GLOSTER: All the events are on that same site.
[00:24:28] 嘉宾卡米尔·斯图尔特·格洛斯特：所有活动都在同一个站点上。
I will be at Vital Voices on October 8th.
我将于 10 月 8 日出席 Vital Voices。
I’ll be all over the country, so please come engage in this conversation with me.
我会走遍全国，所以请来和我一起参与这次对话。
And thank you for making space for this conversation, Nicol.
感谢您为这次谈话腾出空间，尼科尔。
[00:24:39] CO-HOST NICOL TURNER LEE: Oh, always.
[00:24:39] 联合主持人尼科尔·特纳·李：哦，总是如此。
Always.
总是。
So thank you again.
所以再次感谢你。
And for those of you that are listening, please continue to follow our work on the Brookings TechTank blog, uh, which is accessible at brookings.edu.
对于那些正在倾听的人，请继续关注我们在布鲁金斯科技坦克博客上的工作，呃，可以通过 brookings.edu 访问该博客。
Um, you can also look at the Center for Technology Innovation, the work that we’re doing at the AI Equity Lab.
嗯，你也可以看看技术创新中心，我们在人工智能股权实验室正在做的工作。
This concludes another insightful episode of the TechTank podcast, where we make bits into palatable bites.
TechTank 播客的另一个富有洞察力的一集到此结束，我们将片段制作成美味的食物。
Until next time.
直到下次。
[00:25:09] Thank you for listening to TechTank, a series of roundtable discussions and interviews with technology experts and policymakers.
[00:25:09] 感谢您收听 TechTank，这是一系列圆桌讨论以及对技术专家和政策制定者的采访。
For more conversations like this, subscribe to the podcast and sign up to receive the TechTank newsletter for more research and analysis from the Center for Technology Innovation at Brookings.
欲了解更多此类对话，请订阅播客并注册接收 TechTank 时事通讯，以获取布鲁金斯学会技术创新中心的更多研究和分析。