The AI Daily Brief: Artificial Intelligence News and Analysis - A Blueprint for Enterprise Agent Adoption

Episode Date: January 31, 2025

AI agents are one of the biggest enterprise topics 2025, but where does adoption stand? In this conversation, KPMG’s Swami Chandrasekaran breaks down the practical realities of implementing agents in large organizations. From the state of enterprise readiness to frameworks like TACO (Taskers, Automators, Collaborators, Orchestrators), this discussion covers what enterprises need to consider before deploying agents at scale. Brought to you by: KPMG – Go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠www.kpmg.us/ai⁠⁠⁠⁠⁠⁠⁠⁠⁠ to learn more about how KPMG can help you drive value with our AI solutions. Vanta - Simplify compliance - ⁠⁠⁠⁠⁠⁠⁠https://vanta.com/nlw The Agent Readiness Audit from Superintelligent - Go to https://besuper.ai/ to request your company's agent readiness score. The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: https://pod.link/1680633614 Subscribe to the newsletter: https://aidailybrief.beehiiv.com/ Join our Discord: https://bit.ly/aibreakdown

Today on the AI Daily Brief, a blueprint for enterprise AI adoption. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. To join the conversation, follow the Discord link in our show notes. Hello, friends. Today we once again have a slightly different type of episode, but one that I'm really excited about. It is undeniable that the biggest theme this year for most enterprises, or at least the most exciting theme to most enterprises, is agents. I have a whole slew of theories around why I think agents have businesses thinking differently

even more than perhaps some Gen A.I. Assistant-type tools have. But in this conversation, I'm joined by Swami Shandr Sakeran, the head of the USAI Center for Excellence for KPMG. Rather than just a general overview of agents, this conversation comprises a part one of something of a blueprint for thinking about enterprise agent adoption or at least testing. Swami shares his taco framework, thinking about different types of agents broken down as taskers, automators, collaborators. We discussed the most common challenges that he's seeing among enterprises trying to adopt agents, and ultimately we try to provide some positive steps that you can take as an enterprise to advance your agent strategy. We certainly don't get through an entire blueprint for an agent

strategy. We will have to have Swami back to keep going on that. As you'll see, Swami is definitely not your standard consultant. He has a deep technology background, working previously as an executive architect at IBM's Watson, among other roles. He holds more than 30 patents and has authored multiple books and articles on applied AI. And so without any further ado, let's dive into this conversation. All right, Swami, welcome to the AI Daily Brief. How are you doing, sir? Do you go, Nathaniel, thanks for having you over, big fun of your show. Appreciate it. Yeah, we were just joking. So up until about, I don't know, 24, 48 hours ago, we were talking about the hottest topic in AI. I think for a very brief moment that's been

displaced maybe by Deep Seek and R1. But broadly speaking, I think this conversation about agents still is pretty down the middle of where a lot of people are thinking. Maybe before we get into it, though, I would love for you to just share a little bit about what you spend your days doing. That gives context for this conversation. So I live in Dallas. I'm a partner, KPMG. I lead the AI and Data Labs for the firm. And what that actually means is as part of the large transformation program, we're running called AIQ. that Steve Chase runs.

The data, AI and data labs is a pretty significant part. The way I explain my job to my 13-year-old is I do three things. When I say, hi, it's me and my team. We do a lot of experimentation. So for lack of better word, we don't have a full-fledged R&D function at the firm. So we do a lot of experiments, innovation, R&D around things that don't exist today, but it will exist tomorrow, whether it is around how do we use language models or how do we build advanced rag, knowledge assistance techniques or even agentic frameworks or how do we evaluate these models.

The second part of what I do is I help establish standardization when it comes to technology, architecture, and patterns for air across the firm. So we don't do the same thing five times. And the third part of is even my history and being in the advisory side of KPMG, I work with a lot of folks in the in co-incubating new things for our clients. so I get closer to client and I understand problems so I don't get too disconnected from what I feel. So nutshell, I have I think the best job in the firm and a lot of fun and a lot of response to release as well. Awesome. So perfect setup. I think a lot of the conversation today is going to be about the practical, factual kind of where we are with agents and understanding where you're sitting, especially relative to clients is useful. Let's actually start there with that question.

When you think about 2025 as relates to agents, they are obviously a key theme. They're on everyone's mind. But where are we actually when it comes to agent adoption, particularly in the enterprise, right? What stage are we at? And let's start there. And there's a lot of branching questions that I have from there as well. Yeah. Let me quickly set the context, no pun intended, right, to get to every agents.

So when large language models came out, we started interacting with it with prompts, ungrounded interactions, we loved it. And then we slowly started to bring in more context through longer prompts, speed shot prompting and so forth. Then thanks to meta, we have this approach with retrieval augmented generation where we said, look, why don't I intercept the prompts and go to a corpus, bring back the relevant chunks, and give it to the model? So we got our arms and ourselves wrapped around, okay, now I understand the context. concept of RAG or what we call knowledge assistance in KPMG. But still with both of these paradigms, you were sitting and typing prompts. You were away, you were doing it, you may end up doing land chain type chaining and those

kind of things, but you're still typing prompts. There is the action. So agents, scum agents, the whole concept is, can I have these machines go, given a larger goal? can these machines go figure out and plan and go take actions? So whether it is researching on a topic or whether it is reconciling a balance sheet against my ERP systems, it's now starting to do things. So what fundamentally makes agents are how will you define your instructions, your goals expressed as instructions, long form prompts?

How well those prompts are reasoned and understood? into through a planner into tasks that you have to perform. And to perform the task, what tools I need to do the job. Then there are things like knowledge, memory, and context and all bunch of things. So fundamentally, it is giving the large language models, not only additional tools, but the ability to do reasoning in the context of a goal or adjacent set of goals you're trying to achieve. Okay, Swami, you gave a very good theoretical definition.

What does it mean? If you look at what is possible today, all the things I've been explaining are possible in a way using frameworks like LanChain and Lamar Dex and others where you can deterministically chain those steps. For example, if I want to reconcile a balance sheet, I may have two break functions. Each function may have a long-form instruction.

I make that execution of function one in Python. You have that output to section in the second function, and I can achieve it. There's nothing truly agentic about it because you are hard coding the steps. The true agentic behavior is going to be where I express, for example, balance sheet reconciliation. What do I do? As an expert, I say a balance sheet will have these following fields. I look for the following parts in the balance sheet input. Then I go to an ERP system and I do a certain thing.

So you are expressing that as a, like how a human expert will express. The question now comes, can any large language model even reason and understand what you say? Probably till like six months ago or maybe a little before that, they were not. It was very hard for them. Over every iteration of the language models that came out from all the big tech, the reasoning capabilities and more importantly, longer instructions, longer prompts, they begin to do pretty well. Even if you go back to three, two, three years ago,

these longer instructions were impossible to achieve. Right now you can do it. So what you have, the ability is better reasoning, better understanding of what you're saying through these long-form instructions that are very critical for reasons. That was not possible in the past. So what does that leave with us? So you can understand instructions well.

You can break them down into tasks, probably. And it now comes to, can you rely, can you, are those tasks that are broken down? And the tools that are used for those tasks, are they reliable enough for you? The answer to the juries out there. The juries out there in terms of the tools and platforms we have tried and worked with, it requires a bit of handhold. Well, the language models can reason. The act of turning that into a set of tasks, apply, instructions, and to go execute.

It is getting there. It's getting better. But long story short, what we can do today is simple agents. I have come up with a simpler definition or a simple four ways to define the types of agents you can do. Accronoms, TACO, taskers, automators, collaborators and orchestrators, which is multi-agent orchestration. And one thing about TACO is people differentiate between, I've heard people talk about, Oh, certain agents don't get to access all tools. My thing is, in the tachoframer, all the categories of agents, four types of agents,

are going to get access to the same knowledge corpus. It's going to get access to the same breadth and depth of tools that the agents would need to create actions. It will have access to memory. It will have access to the same algorithms. So all of those four are fixed. So what is different? The difference between the four comes down. to planning an orchestration.

The T and TACO taskers, there are singular goals, one goal, but can break down to multiple tasks. They can be chained, easy to manage, easy to test, easy to roll out. When you go to Automators, which is the next, they typically go to cross-system, cross-application. So these are end-to-end processes,

order to cash, lead to cash, procure to pay, hire to retire. They touch. multiple applications and multiple systems. So the goal may be similar meaning ensure streamlined order to cash process execution, but they break down to sub-goals. Each other sub-goals may touch different applications and different systems. So it gets a bit complex in terms of the scope of what it does. Planar gets complicated, orchestration gets complicated. In the orchestration, you have to manage state and all these things.

The third part is collaborators. This is where I've been pondering over the question. So there is this concept of can AI be used as teammates, agents be used as teammates. They're no longer you're telling the agent to do something it comes back. You work with it. It's like how you work with your team member on a daily basis. There's more skewness towards human collaboration, partnering with the machine to get things down.

It's there in the other forms of agent, but it is even more so on this is just predominantly burdened. And the last bit O in the taco is the multi-agentness, where I have agents calling other agents, there is interagent collaboration. Of course, the complexity becomes more with all this. So like I said earlier, where are we today? I think there have been a lot of experiments, prototyping, done with the taskers,

people are inherently because there are quite a few platforms, open source commercial included, where you can build them quickly. And we can talk about that. But I think those are, in the year of agents, if 25 days, I would see more taskers. That's my prediction. Do you think it's obviously very dangerous to sort of prescribe one right path without the context of any given organization. but do you think that that that toggle framework actually is basically are they four separate categories only or do they have some sort of linear relationship with one another as you're thinking about adoption if you're sitting in an enterprise where you know it makes sense to start with taskers and then move to the next or you know how do you think about that yeah this is not I don't want this

to be a contrived framework where we retro put everything into one of these four the framework is meant for a mental model mental picture look how can I break down easy? Not everything. Because the reason for this was everybody jumped into multi-agent coordination without even thinking about the basics. So that is what. Second is more than likely when you go talk to clients, they're going to talk about scenarios which will not only overlap,

but will require their focus maybe more than likely starting with, okay, let me do end-to-in-off process automators. Because that's where I need. I want to streamline my store performance management, or I want to streamline my procure to pay process. Or when you go to another client, may say, look, I'm more focused on augmenting my human potential, so give me an AI agent that can act like a teammate for my AP, AR, ERP, finance kind of domains.

So, yes, it is dangerous to put things into the bucket, but that's not the point. The point here is to demystify the whole agentic system and how complexity comes. And if you start to amalgamate and combine, that's okay. but at least you understood the individual feedbacks. Yeah, it's interesting. You know, I think that one of the things that makes agent adoption fascinating as compared to, for example, sort of broader Gen. AI adoption over the last couple of years, enterprises moved very quickly relative to previous,

you know, technology changes to grab onto Gen. I and try to sort of harness it. Now, obviously, there's still tons of organizations that are behind, feel behind. Very few organizations, I think. we tend to find that the organizations who are the farthest ahead also have the greatest awareness of how much more they still have to do when it comes to adoption. So it's not like they're sort of, you know, at the end state or anything. But I do think that because they've been watching agents come down the pipeline for a little while, they maybe have a stronger sense in general

of how they want to eventually use agents, the possibilities that have them most excited. And I think that it, you know, it might be leading to some of what you're seeing around they're jumping to sort of exactly what they would like out of an ideal state of what an agent can do. They're imagining even ahead of where the technology is rather than sort of just racing to catch up with what it can do now, which can create challenges just based on what's actually ready for prime time and what's not at this exact moment. Yeah, everybody has an expectation and a notion of what this agent should be for them. If you go look at the customer service and marketing function, they say, my version of an agent is

can I put a digital version of a customer or a software development or sales development representative and it can talk to clients it can ask the questions it can help close a sale and get paid a commission

and come on so they start thinking it like synthetic employees you go into the enterprise you go into the mid and back office functions they think in terms of processes There is a particular way in which I receive, review, approve or deny invoices as part of my larger procure to pay end-to-end process. So I have a conception of how agents should be in that particular way.

It is not once I switch all, like you said, but at the same time, the key responsibility is when you go talk about them, you're not trying to take an existing technology and retrofit and say, oh, I have agents. So as an example, one belief I have is good old business process engineering, like how you sat and designed business processes for inton processes. It took a particular approach. Process engineering came out where he said, decompose your domain, break them down at a level once through level N,

could go up to level seven and eight, where you kind of have a massive swim lane view of how your process looks like. That's how we represented processes. That's not how machines think. Now with the reasoning capabilities, I could express that same thing, almost like a long-form instruction. And you leave it to the machine to say,

look, you go to find the process, the steps that are needed to execute it. So there is also a change in with how we approach designing the agents that is also essential and important. The outcome is the same. I want a better, efficient, leaner process. But you're approaching it in a different way. So the point being, the entry point for agents are different. They're all going to converge at some point in time,

but given where we are and the stage we are in, the expectations are widely done. Today's episode is brought to you by Vanta. Trust isn't just earned, it's demanded. Whether you're a startup founder navigating your first audit or a seasoned security professional scaling your GRC program, proving your commitment to security has never been more critical or more complex. That's where Vanta comes in.

Businesses use Vanta to establish trust by automating compliance needs across over 35 frameworks like SOC2 and ISO-27-01. Centralized security workflows, complete questionnaires, up to 5X faster and proactively manage vendor risk. Vanta can help you start or scale up your security program by connecting you with auditors and experts to conduct your audit and set up your security program quickly. Plus, with automation and AI throughout the platform, Vanta gives you time back so you can focus on building your company. Join over 9,000 global companies like Atlassian, Quora, and Factory who use Vanta to manage risk and prove security in real time.

For a limited time, this audience gets $1,000 off Vanta at vanta.com slash NLW. That's V-A-N-T-A.com slash N-L-W for $1,000 off. If there is one thing that's clear about AI in 2025, it's that the agents are coming. Vertical agents by industry, horizontal agent platforms, agents per function. If you are running a large enterprise, you will be experimenting with agents next year. And given how new this is, all of us are going to be back in pilot mode. That's why Super Intelligence is offering. a new product for the beginning of this year, it's an agent readiness and opportunity audit.

Over the course of a couple quick weeks, we dig in with your team to understand what type of agents make sense for you to test, what type of infrastructure support you need to be ready, and to ultimately come away with a set of actionable recommendations that get you prepared to figure out how agents can transform your business. If you are interested in the agent readiness and opportunity audit, reach out directly to me, NLW at B-Super.AI, put the word agent in the subject line so I know what you're talking about, and let's have you be a leader in the most dynamic part of the AI market. Hello, AI Daily Brief listeners. Taking a quick break to share some very interesting findings from KPMG's latest AI quarterly Pulse survey. Did you know that

67% of business leaders expect AI to fundamentally transform their businesses within the next two years? And yet, it's not all smooth sailing. The biggest challenges that they face include things like data quality, risk management, and employee adoption. KPMG is at the forefront. of helping organizations navigate these hurdles. They're not just talking about AI. They're leading the charge with practical solutions and real-world applications. For instance, over half of the organization surveyed are exploring AI agents to handle tasks like administrative duties and call center operations. So if you're looking to stay ahead in the AI game, keep an eye on KPMG. They're not just a part of the conversation. They're helping shape it. Learn more about how KPMG is driving AI innovation at

KPMG.com slash US. What do you think about as you're advising clients or even just thinking about it broadly and you're thinking about agent readiness in the enterprise? What are the some of the pillars of consideration? How much is it about data? How much is it about policy? How much is it about understanding objectives as you've just articulated? What are some of the key pillars of agent readiness?

Yeah, you kind of gave our three out of the, things I was going to say anyway. So first of all, why agents? I start with that question. What is the rationale? What is the motivation for you? So first, define, don't go go to technology called agents yet. What is the problem you're always trying to solve?

So if I'm a client, if they come and say, if they're a retailer, they come and say, you know what, I want better top line growth increase in my stores. me put in my brick and mortar stores. Okay. What are you doing today? They say, okay, I have these things, but stores, the sales get affected because certain stores don't follow certain kind of policies and procedures. They don't take into account customer satisfaction or customer reviews and all those kind

of things. Okay. Then we go in and say, okay, the goal and objective is to have a better, more tangential approach how do you do store performance analysis so you can improve the performance and increase your top line and do. So number one is what I're trying to do and is agents even the right answer. So let's assume you've gone down the path of saying, look, I want to optimize my processes, reimagined my processes at the same time optimizing my human resources. Then you talk about, okay, where is the data coming from? Do you have the data? Do you have access to all of the data? Have you been, first of all

instrumented the data if it is if it has to be digitized and is that data made is clean and it's all the good things about data availability and readiness and everything the third one is i i don't think you you mentioned in the in the list nathaniel which is who is the human expert who can articulate what is happening today and what needs to change how are we going to elicit that knowledge whether you pick a domain you pick any any intense domain if it is some even if it is some even if it something as simple as customer service. From the point a customer comes and raises a request for refund, what do you do? What is the process you follow? And what is the way to reimagine from that point onwards using agentic concepts? So human expertise is still needed to articulate.

I mean, there are theories floating around. Can I go do simulation? Can I look at what humans being doing and learn from that? Yeah, you can, but they're not fully reliable yet. So why agents data, human expertise articulating the whole thinking process and the process, and how agents have to be built, then getting into policy three that things, okay, are there things you, how much of autonomy you want to give to these issues? It's not a, it could be at a very broad stroke principle level saying, look, I don't want any decisions that have a financial implication to be approved without human in the law. maybe I want three steps of three stages of human in the loop. So there is a whole strategy around how do you bring in humans, where do you bring them in, where is the level of oversight, what does the kill switch equivalent for agents look like,

what do you want to stop agents for a day, what is your fallback mechanism in case these don't start to work? So all of those policy, trust, security, reliability aspects is one big bucket. The fourth important market is everybody and the same. is a very opinionated topic I've seen with clients is how are you going to build agents? Okay, everything fine, you got the data, you got the experts, you've got policies, you know how to build him, where are you going to go build it? So today, there are a dime a dozen open source frameworks, the big tech, smart tech startups, they're all, they all have their platform. So where do you go standardize and build? Again, my thought process there is till this whole thing

settles down, you may have to remain polyglart and bigger. a few choices, be very opinionated and go build and try them out. And some are going to work, some are not. So you have to be ready for consolidation and merging. So what is the tool, technology, infrastructure that you're going to go to? I'm not even using LLMs, I'm assuming they're going to get awesome. They are awesome already. They're going to continue to get awesome.

And the last bit of their own skills, do you have the skills to build this? And one more thing after this. Okay, you have the skills. Building agents is one thing. the day two plus operations is a completely different thing. How are you going to sustain? So we've talked about model drift and data drift. Now comes Asian drift.

What's the guarantee the agents are not going to drift? It's going to deviate away from what it was built for. How do you keep them, upkeep them? Is the data changing? How good of a feedback are you providing back to it for reinforcement? Those all come in the day two plus operations. So top of my mind, I think these are these are the kind of categories of things that would look at. And do you have, I think it's a really useful framework.

How much do you think people, how much are you seeing people's first experiences being something that they're kind of rolling their own, you know, with one of these general frameworks versus trying something that's more off the shelf? I mean, this is kind of only a question for the last few months as more off the shelf things have been available. but, you know, working with a customer service agent, or does this have to do with which category of agent to use your framework they're actually thinking about? Yeah. So if you double-click into where are you building agents, I think it double-clicks into three sub-questions or sub-areas.

Are you going to build your own using open-source? Are you going to pick a commercial platform like a co-pilot studio or agent space? A third option is, are you going to buy the agent? So you go to Agent Forrest is going to say, okay, I already have a sales coach agent. You're just going to buy it, configure it, and use it. The experience is changing by the month. What we have today is not what we had six months ago. Again, there's another one.

The way I look at the whole agentic tooling spaces, there is low-cord tools, like Copilot Studios and those kind of the work. Then on the far right, you've got the pro-cord tools like the, the Langer apps and crew AI and autogens of the world. Then in the middle, I call them mid-code. You can go back and forth. Meaning, I can write code, I can write in Goey, drag and drop, so I can do both. Initially, people tend to go use the pro-code options,

and they realize, while it gives them a lot of flexibility, they have to end up building a lot of things down on their own. So there's a lot of lines of code to write and maintain and manage. brittleness starts to kick in unless you have a well-coordinated engineering team development team you may end up recreating the same thing

for example the same tools to do the same thing may get recreated multiple times so there is that risk of having and you need to have a special set of skills and capabilities to do coding by yourself now if you come to low code

I mean I could get started quickly very easily but but I've seen roadblocks where they say, oh, I want to do this Excel comparison for one of the steps in my agent, and I cannot do a very deep Excel analysis because my Excel has multiple complex cells and rows and headers. As an example.

Like I said, that's why the whole polyglard approach is needed. Like, you need to first decide what is my agenting architecture going to look like? What are the tools that I need as of enterprise? Let's go figure out the strategy to build those tools that are reusable way. And then it doesn't matter if I'm building my agents in my throw code or on my low code, they all access the same set of tools. So let's focus more on getting the task done with the same set of guidelines principles and safety. And if you are ready to keep upkeep these agents from day to onwards, you make a choice. So I think the jury's out there in terms of not one

platform is got everything you need. If you have something, then there is going to be something it does not give you your reflection point you get. This is a, I don't know if I'll phrase this question right initially, but, you know, with Gen. I right now, sort of non-agentic Gen. AI, LLMs and, you know, assistant copilot style tools, a lot of adoption is happening, at least mediated by some central body in the enterprise that's tasked with thinking about AI transformation, right?

So maybe it's a repurposed innovation group that touches all the lines of business and all the back office functions and all the things that sort of just understands everyone's different stakes and who become the conduit for different use cases and different tools and things like that. So it's top down, not in an unaggressive kind of way, but in a, you know, still, still like coming through a central entity. Do you think that agentic adoption is going to mirror that? It's going to come from central groups analyzing all the different options, or is this going to be a little bit more bottoms up where it's a specific department or a specific line of business or a specific area, you know, experimenting with something that's direct and purposeful for them.

You cannot stop innovation in the grassroots. That's the reality. People are going to keep innovating and coming with new approaches. Because the role I'm in, I belong to that central organization. So fair disclosure, right? I'm providing my perspective with that sitting in that part, in that side of the world, I believe helping standardize on the approach, the technology, the platforms, including safety that you incorporate when you're building agents will go a long way in helping folks in the departments and different business units, spend their time and energy in building. Where I see a lot of time and energy being spent is trying to build your own agency platform.

or trying to make your own agentic platform. This is like saying, I'm trying to build my own, I'm trying to build a car, but I have four groups in the company, and each one of them is building their own supply chain or the assembly line. Why even try that?

Why don't we build one good, efficient, morality, Toyota, Tesla, you pick the best supply chain, for the assembly line, including the supply chain that powers it? And you focus on, designing the Model 3 or the Toyota Camry or the whatever your favorite car is. So standardizing, giving them the platform and providing the guidelines and let them

focus on the hard part. The hard part like I was telling earlier, eliciting knowledge of everyday work in translating that into an agent. That takes time, that significant piece of work. So who's going to do that if everybody's focusing on? I'll also build the platform. and I will also build the agent. So it sounds like a bit of a both-hand. There's going to be functions that are relevant for kind of an org-wide or at least cross-functional discussion from an infrastructure perspective in particular, while there's also a clear kind of purpose for what the individual units or groups are going to actually need and understand.

Yeah, yeah. And one other observation data point is we're already finding the individual groups heavily time constrained, meaning they don't have a lot of time to go to R&D, pick a paper, platform, evaluate a platform, evaluate choices, what kind of evaluations do I do on agents, this versus another. They already kind of, they already have like things to go ship and build. So trying to take this as well as those as much as away and have the central group help provide that guidance. Let me go a down even a level from, from that sort of department or functional or group level. How much are you thinking about?

individual level, employee level adoption and the challenges they're in, either when it comes to, you know, getting employee perspectives on which tasks are actually suited for automation or which things they, you know, they'd like to have agentic support for, as well as, you know, a question of employee attitudes and concerns around, you know, replacements and things like that. How much are you seeing that enter the discussion as companies are moving into this space? So in one side, there are tools, for example, KPMG's role. Microsoft M365 copilot to all of our employees in the US, for example, except for federal. So they have access to all of the tools, the ability to create what are called personal co-pilots

where you can point it to your own SharePoint corpus and start to interact with it. So they can pretty much do this in a matter of a few seconds today. So there is that level of capabilities that are made available by big tech like Microsoft and made available for large corporations. The reality is they are made available. They are there. The next evolution in that is they're also going to say, OK, you can build your own agents to automate your daily tasks.

So there is one theory from the big tech where they want to push the tools for more adoption, better adoption. They're saying, look, you can build assistance agents on your own, and it's going to be easy. My take is, look, well, there's, well, That is all good on paper, but imagine you're going to have hundreds and thousands of these agents all over the place. The kind of actions the agents are going to take, we have to carefully manage them. You don't want it to start doing things that will leak your IP, leak your knowledge, leak your data, put your at risk and be one.

So one tool is people who are builders, the builders of agents will have to be certain types of people who have gone through not only skilling, training, and other kinds of things, also understand the implications of building agents in a particular way. So you're going to start seeing personal agents that is confined only what I do as work. So today in my computer, I could have a shell script that can do things that is confined to what is happening in my own specific environment. Enterprise grade agents, I think, will take a path where it will be built by folks who have gone through a certain level of pedigree instead. If I could say that.

I don't think either of them are going to stop. Do you see a convergence of those at some point where companies start? I mean, one of the fascinating things about Gen. I in general is that it's the first time that Shadow IT has been, while yes, a concern, also an area for innovation that they're actively trying to understand so they can potentially bring in, right? Like, you want to understand what people are using their personal g-mails for to sign up for, not only because you want them to not put important company data on those platforms without your knowledge, but also because you might want to adopt those. And given how much of a race there is

to the personal assistant side of agents, right, we're recording this just a few days after operator has come out, I can see there being a sort of a blend where enterprises start trying to adopt agents from a top down kind of way, or at least sort of a unit by unit, group by function by function kind of way. And employees are bringing in assistance that have sort of started to automate their own personal processes at the same time. Yeah. Since the birth of operators, let's take that as an example. I could build, when operators may come available for everybody,

I could build an operator that I could use for my, for example, my weekend planning or my calendar, assuming I can log into Outlook on the web, look at my calendar and see overlapping meetings and come and tell me which ones I should consider canceling as an example. But that's me having, unleashing an operator, building and unleashing an operator that is happening in my personal environment space.

Assuming 10 other people find about it and say that's a very good use of operator, that's a very good personal agent. Can you share that with me? So the point I'm trying to make is the personal agents, the scope of sharing, is going to be limited. If you keep it that way, it's not permeating across the enterprise. It's still being built on. This is not like somebody gone.

and built their own agent on an unapproved platform. I'm still talking about approved platforms, but built personally, but the scope of sharing is limited. I foresee a world where you're going to see organic innovation happening and somebody's going to crack the the nut on, oh, this is the most innovative use of operators or agents or co-part studio or whatnot,

that I think should be made available at the enterprise level. To go through that level, you've got to go through stage gates of testing evaluation, safety, and other things. So you have proper governance in place. Because for the enterprise, I see them no different than treating them as products. They're rolling our products in your enterprise. You're not just going to roll out things randomly on the fly without knowing what it is doing in your enterprise. So I think the, I had, I had some idea coming into this what I wanted to do. But what's become clear is that I think this episode will kind of stand and I'm going to frame it as,

almost sort of like an agent readiness checklist. But I think we just did part one. What I would suggest is maybe one wrap-up question, but then we should come back and do this, you know, maybe next month and do a part two where we get into maybe some more specifics around use cases and things like that. I guess until we get there,

if you had one general piece of advice for the next month, you're not going to get to talk to these listeners as they're thinking about adopting agents in their companies, what's one thing you would encourage them avoiding or trying or just setting as part of their framework to kind of maximize how they think about adopting agents in this year? Yeah, one thing is always hard, but let me try. One thing I'll highly encourage is don't stop experimenting.

I mean, you have to do that only then you would understand what is right or wrong. But one thing I would highly encourage everybody to go do is talk to your respective transformation, technology AI leaders. First question to ask is the things I've been talking about, what are we going to do? If I have the next best agentic idea, where do I go build it?

Where do I build it in a way that it is not throw away work? Because that could be a rallying point for many things, meaning, what are the agents? What kind of agents are they? How do I build those agents? What data do I need to build agents?

Because I've seen everybody talk about thinking about agents talking about agents, debating about them. But when it comes to rubber hits the road of I need the data, I need to go build them, it becomes analysis for analysis. So we are in the mode, if we are very, if all of us believe this is the year of agents, then you should have already picked a platform. If you have not, highly encouraged, go, think about where do you go build? And then everything else would follow. Are you, are you, are you, do you have the skills to go build it? What else do you need to think about? They all naturally follow.

Awesome. Well, like I said, I really do think this should be a part one and we should come back again. But I appreciate you spending some time with us today. I think it's, you know, everyone is trying to wrap their head around this particular question right now. So invaluable to have you here to talk through it. Thank you, Nathan. Happy to come back.