Goldman CIO Marco Argenti on the Warp-Speed Improvements in AI

Running a business means dealing with a lot of overly complicated software, and most CRMs tend to follow the same pattern. They're packed with endless features you'll never use, interfaces that feel clunky, and teams end up spending way too much time just trying to find basic information. Today's sponsor, Pipedrive, is a simple CRM tool designed for small and medium businesses. Pipedrive brings you entire sales processes into one dashboard, giving you a crystal clear, complete view of sales processes and customer information, designed to help teams stay in control and close more deals faster. It all centers around the Visual Sales Pipeline, where you can see every deal, what stage it's in, and what needs to happen next. Since everything is in one platform, Pipedrive is designed to unite your team, keep track of sales tasks, and stay on top of your leads. Switch to a CRM, built by salespeople, four salespeople, and join the over 100,000 companies already using Pipedrive. Right now you'll get a 30 day free trial, no credit card or payment needed. Just head to pipedrive.com/simplecrm to get started. Thanks for listening to All Thoughts. Follow the show on Amazon Music for more future episodes or just ask Alexa, play the podcast All Thoughts on Amazon Music. Bloomberg Audio Studios. Podcasts, radio, news. Hello and welcome to another episode of The All Thoughts Podcast. I'm Tracy Allaway. And I'm Joe Weisenthal. Joe, the thing about AI. I feel like it's accelerated all of our timelines. It's phenomenal to me to think back to the days of chat GPT. When did that come out? 2020. 2020. 2022. That's just crazy to think. It's really unbelievable, the gap. I've been thinking about this, like just the explosion of capabilities. Yeah. And the thing I've been thinking about is that after the first year or so and it came out, we talked to executives and it's like, how are you using AI and your workflow? Everyone's experimenting. It's great. Everyone is using Chat GPT. It's always very vague. And now in 2026, the story is that AI is so powerful that it's going to destroy all these legacy software companies. So what I would say is we must be past the age of experimentation. I think that any company-- Yeah, really use cases. Yeah. --using it better have some example of like, here is a workflow where we're using it. Well, exactly. And to this point, now that we're past the age of experimentation, I'm very curious how executives and managers are actually evaluating the return on investment in AI and what they actually want to see from it at this point. So, you know, are you going to replace all your third party SaaS contractors with internal coders? What does that look like from an actual head count perspective, from a cost savings perspective? We can actually get some concrete details on this now. So I'm very excited to say we do, in fact, have the perfect guest. Someone who we had on before to talk generally about AI and someone at a company that has been doing, you know, they got into it pretty fast. The last time we spoke to this person was in 2024 and even since then-- Eight years ago. I guess it just feels like light years. You know, in AI time. So just one last thing on the last few years of AI, which is that when Chagy PT, when it came out, I played around with a lot when I had it right poems and all this stuff. And then I bet if you actually looked at my AI usage, went for through a trough, whereas like I wasn't really getting any productivity. There was nothing it really could do that I needed. It still sort of seemed like a toy. So I had this, like a tense burst of use for the first several months. And then this trough, and now these days with the expansion of capabilities, particularly cloud code, I'm finding all kinds of new things. So there is like, we're out coming out of the trough. I think a lot of people are actually finding things, at least, if I can generalize from my own experience. Yeah, absolutely. So we do, in fact, have the perfect guest. We've brought back Marco Argenti. He is, of course, the Chief Information Officer over at Goldman Sachs. Someone we had on the podcast back in August of 2024. So Marco, thank you so much for coming back on. Thank you for having me. How much have things changed for you? Does 2024 seem like 20 years ago now in AI time? Yeah, I barely remember even what happened back there. That's a nice way of saying you forgot what we talked about on the podcast. That's fine. Yeah, maybe that. But literally, like, things are really changing on a weekly basis almost right now. And if I look at the evolution, not only since a year ago, but even since six months ago, I think it has been nothing short than a revolutionary. A year ago, we barely talked about agents or the word almost didn't exist. We were using AI as like a chat companion. Yeah. There was a search function. Yeah. It was telling you, oh, I'm sorry. I don't know who's the president of the United States because my cutoff date is like a year and a half before or thinks of that nature. Now you can say, hey, as a person, you can say, hey, my plan just got canceled and it's going to redo all your plans. It's going to check for like available flights. It's going to do all these capabilities of personal assistant. And that translates in corporations also in a lot of utility that you can see in everyday tasks. So I would say, you know, what I like to say to my people about us in general, I say, this is not a drill. This is real. You know, it's not the age of experimentation anymore. There's a tool that now can do a lot for you. And so we put it to work. We put it to work starting from developers, but don't expanding in many, many other areas. So I would say actually, if I look at the increase of capabilities of these models, what we've seen in the last six months or so with really the evolution of this advanced reasoning capabilities that came out, I think that finally got us the confidence that you can use AI's for everyday work with the right supervision. And also in many cases for mission critical applications, it's not a toy anymore. It's something that you can expect results from. And I think that's the biggest change. So today, I would say that there is almost nobody in Goldman that is not touched in a way or another by AI's. We gave our GSI or GSI assistant to 47,000 people. Most of them use it every day. Most of them use multiple times a day. And what's interesting is it's like the first time you see a tool like Microsoft Excel, you can almost not predict what people are going to do with that. Maybe it's born for doing some form of accounting and then people write entire applications on top of that or use it for project managers, management or things of that nature. And AI is kind of turning that way. If I look at what people do with that, it's really things that surprises every day because of what do you give us some examples? So in production right now, what are some workflows or novel things that were not workflows before the UC within Goldman that AI is doing for people today? So let's start from the GSI assistant. That can answer really complex questions based on external and internal data that generally before used to take sometimes hours or even days, sometimes weeks to answer. It can do very complex research for you in topics. For example, we can ask questions that come from clients such as, hey, how does the recent geopolitical events on the Hormut's trade actually impact the portfolio? What could be a potential rebalancing strategy? Or you could ask intersection of like, I don't know, how does the certain fed decision on interest rate actually impact the volatility of certain assets? So you ask this multidimensional questions and what GSI assistant does calls out a model, retrieves the relevant information, creates a plan to answer that question. That's kind of the key because this AI is the really plan before responding rather than just giving you the first thing that comes to mind. And so that's what kind of at the very surface thinks that one of the most common use cases which is we really enhance the client experience by being able to answer questions internally and externally in a much, much faster way, but really complex questions, not simple questions. We had to wire up hundreds of data sources and also most importantly, which is something that I tell everybody that asked me, hey, give me some advice on how to implement AI in a corporation. Data quality is really the determinant between good AI and not so good AI. So we do a lot of work to not only take a bunch of data, but also making it understandable to the AI. So for example, just to go a little bit deeper, we have a tool called Legend AI, which is our lake house, which allows you to go from query to MCP server connected to GSA assistant, i.e. from data to answers. You can wire it up literally in two or three clicks. And it does all of that for you. So the quality of the data, the quantity of the data, not only that is not just the bitter lesson here, but it's also the lesson of you need to curate your data. You get better answer disproportionately. That's something that has driven that. So that is kind of the knowledge aspect of AI, which is, I would say, the most widespread because every single one in the firm has that. And it's the highest users. We are like way above a million prompts per month and it's growing really, really fast. And then of course, you're asking me like real impact in production. Every developer in Goldman is enabled with the agent AI. So we were probably one of the first, if not the first to launch a dev in almost a year ago, which is the fully agent developer assistant. We have Cloud Code, we have many other tools, GitHub. copy, that agent, et cetera. But on that, you really see the step change. There is no question that there is change in the way developers work. And by the way, it's not just about doing the exact same things more efficiently. It's changing the way developers actually do their work. And there is very, very easy to see how that kind of changes the paradigm of what a developer does. You're much more of a product manager, you're much more of a planner, you're much more of a-- idea generate. The most important thing for a developer today is to be able to explain things rather than jumping into other things. I don't know, you want to-- No, I just going to say that resonates because I've been like vibe coding, but I can't explain how any of it works. So if someone is like, you know, I build little toy apps and stuff, but I get really anxious. I couldn't explain. That's why I'm not a software developer. Well, just on this note, I mean, people tend to talk in generalities when it comes to AI boosting productivity or maybe AI changes the way we work or leads to some new ideas. From your seat at Goldman, your manager, you're looking at the bottom line of all these businesses, what exactly is the outcome, the specific outcome that you would like to see from your developers using something like Cloud Code? It's really about increasing the output. So I want to see-- I was actually having this discussion this morning. I was looking at some of the reports and some of the deliverables for our Cloud migration, which is a very important thing for us. And I was looking at this really big project that was saying, it was not only green, it was like two months ahead of schedule. And I was saying, this is how we know when things are going to work. You're going to consistently start seeing projects that are actually finishing ahead of schedule, which means that then people are ambitious. They want to do more. And therefore, you end up with output that is much higher than what you had before. And listen, with developers, obviously, the biggest question that everybody asks is, OK, what are you going to do? Are you going to cut developers this and that? So first of all, with all the innovation that I've seen in the last three years or so, I can never see in a moment where really people were reducing the number of developers. Because if I look at the things they were not doing in a certain year because of budget reasons, because of complexity reason, because of prioritization, the stuff that is below the cut of the backlog, it's a lot. And a lot of that is really driving the growth of the business. So it's good to have the optionality to do it. You have the optionality of saying, I now have 120% of my capacity. I have 130% of my capacity. Do I want to do 130% more? Great. If I don't, I have the option to reduce. So that's really how we measure it. It's really the impact on the timelines of delivery. It's output. It's basically quality and timeline becoming quality gets actually better. And the timelines get short. So that's what we measure. [MUSIC PLAYING] Running a business means dealing with a lot of overly complicated software. And most CRM's tend to follow the same pattern. Pipedrive brings you entire sales processes into one dashboard, giving you a crystal clear, complete view of sales processes and customer information designed to help teams stay in control and close more deals faster. It all centers around the visual sales pipeline, where you could see every deal, what stage it's in, and what needs to happen next. Right now, you'll get a 30-day free trial. No credit card or payment needed. That's pipedrive.com/simplecrm. Hello, I'm Stephen Carroll. I'm in Brussels, where many of Europe's biggest decisions get made. And I'm Caroline Hepgit in London, with the hosts of the Bluebeg Daybreak Europe podcast. We're up early every week day, keeping an eye on what's happening across Europe and around the world. We do it early, so the news is fresh, not recycled, and so you know what actually matters as the day gets going. From Brussels, I'm following the politics, policy, and the people shaping the European Union right now. And from London, I'm looking at what all that means for markets, money, and the wider economy. We've got reporters across Europe and around the globe feeding in as stories break. So whether it's geopolitics, energy, tech, or markets, you're hearing it while it happens. It's smart, calm, and to the point. And it fits into your morning. You can find new episodes of the Bluebeg Daybreak Europe podcast by 7am in Dublin or 8am in Brussels, Berlin, and Paris. On Apple, Spotify, YouTube, or wherever you get your podcasts. Obviously, one of the big questions for the market this year is what is the impact of AI on legacy software providers? And there's various theories about how they could be disrupted. There are reports I think about Anthropic having quote, "Forward deployed engineers inside Goldman Sachs." So Anthropic employees billy out AI systems internally. Maybe that could replace some legacy software. Right now, can you say, like, there is a change in the balance of power when there is a given piece of software for negotiation? I think generally there is. OK. First of all, there has kind of always been that tension because, imagine, for example, imagine when there were software that didn't run on the cloud. And then all of a sudden, a bunch of new vendors are coming to USA. Hey, wait a second. Why are you running that on your mainframe or your on-prem? Why don't you run it in the cloud? Or remember, when software needed to be installed, then everything became browser-based and so there has always been a little bit of a cycle and a renewal. What I say is that today, that cycle of renewal is much faster. That's really what it is. And I would say, I generally resist making, like, really broad categorizations. AI is, by the way, the most possible. Sumi is like saying computers. But even software is very broad. And so within the software category, I think there are winners or losers. There are winners in the long term and losers in the long term. But it's really like people tend to make it a category, and then maybe throw the baby with the bathwater. So here's an example. To me, the question that I ask myself with regards to which vendors are I going to-- which software am I going to have like a few years down the road-- is software generally is attached to a process or a certain ways of work. It does something for you and it puts it in the form of an application that you use. The real question is, is that process and/or ways of working going to be the same? Or is it going to change in five years? Then you can determine what is basically the likelihood that the software is going to be robust to that or not. Example is accounting or closing the books. Going to be very different five years from now? I don't think so. Really. It hasn't really changed. I mean, everything changes, but it hasn't really changed much. So if you're operating in the general ledger type of category, I don't think all of the sudden you take a GPT or a cloud and it's going to close your books magically. You still do the accounting and the security. You still need to do a lot of that. And it's very regulated, importantly. It's extremely regulated. Jurisdiction by jurisdiction, country by country, product by product, industry by industry. So that part is kind of to me in kind of the safe mode, and then you go to the other end of the spectrum and you have sometimes software that is aligned to the way people do things today, like software being one of them. If you look at the software developer lifecycle, a lot of that is changing. Developers are developing software by developing specs today. And so if you're too much in the weeds down there in that mechanics and you don't adapt for AI's or agents doing that work, a software development lifecycle, deployments, rollbacks, monitoring, observability, and all that, I think that part will be very much disrupted. Or if you're adding a sort of a UX on things, that's another class. You have a very simple process, I don't know. You're doing surveys or expense reports or whatever. And now people are going to start expecting their personal assistance or agents to kind of do all that mechanic for them. And so I think that part is probably something that has a bigger question mark on top. And so I always ask myself for the process question first. The process transformation section, the question first. And then consequently, what's the tool that is going to support that? Just a press on this point, though. Have you replaced any third party software providers with something that's been developed internally through AI? We have terminated contracts already, yes, absolute. No, I'm not going to ask you your follow-up questions because I'm not going to pay your names. I'm going to take care of some of the software stocks that you did. But overall, yes, absolutely. And the whole buy versus build, the equation has changed quite a bit. Buy versus build was always like, OK, guys, how long does it take to do this? And you get an answer, which is, well, we can do it in like X amount of years and X amount of millions of dollars. Now I'm starting to see people coming to me and say, by the way, I had some time this weekend. And here is a perfectly working application. So the cost or at least for simple applications, the cost of kind of build. from a time perspective, from an extra cost perspective has gone down quite dramatically. So right now, the little things are most likely going to be built. The very big large software that is to be deployed that scale across thousands of people and etc. said that big complexity, as you know from we all do toy stuff with our clock codes at home and whatever. You know, there's still some rough edges, you know. And so it's harder to think that all of the sudden the big applications are going to disappear. And so that's really what what I'm saying that if I look at the applications that I buy today, there is a lot of small applications. And so the build is kind of the pendulum is starting to swing back towards the build, at least for that category for sure. What is a forward deployed engineer? I know that's like one of the hot buzzwords of 2026 and I saw headlines that there were anthropic forward deployed engineers at Goldman. I have no idea what that means actually. What is that term? What do they do when they got there? Okay, so I think listen, that name has changed quite a bit. Oh yeah, it's already used the out of date terminal. No, no, no, no, no, no, no, that is the latest. Oh, you are fully you're in the V in the V latest of that term. But remember, I mean, there was a time where you used to call them solution architects, right? And so the point is right now, I think one trend that I see, which is also kind of true for us, is when things change so much and so rapidly, you kind of want to go to the origin of who produces this new thing. So the least intermediaries you have, and probably the faster you can go. So going and working directly with the model providers is generally good idea. Because if you're putting someone in the middle, this company is going to have to be trained, there's going to have to be, you know, there is a cycle, which at this point of very rapid change is going to slow you down. And so the first thing that that term means is those are people that are actually normally building the product. They're normally building the cloud or GPT or X or so. That's the first differentiation. They're straight to the source of the AI production in a way. And second is that they are generally product people. So people that have actually built those tools. Rather than people that are more like support and deployment. Okay. And so this characterization is you take the classic cell support team or solution support team which was mostly doing integration. And when things are so rapid, it's like, you know, imagine if there is like something like, I don't know, if a closed style works, it changes so fast. Instead of a fashion assistant, you want to talk to the tailor because they can actually make it. Things are changing so fast. So those are the tailor's best. So on this note, one of the things we heard in support of SAS was this idea that while integration is still going to be really important. And that's really going to be like the major hurdle for a lot of the stuff. Have you found that AI is making integration even faster at this point? Has that basically become irrelevant nowadays? No, I think integration is extremely important, especially for the industry cause like systems of record. So when you do something like, you know, when you do a process, then you have a source of data like your CRM systems could be a system or record or you have your client system or record your accounting system or record. And those when they become the authoritative source of an answer, they need to integrate with the rest of the firm. And the rest of the data, the rest of the application says, so I can see that those vendors that sit on top of those, they can argue that they will implement, there's nobody that is better positioned than them to implement AI, they will kind of reach outwards and actually do that kind of integration. So I think those who will evolve so that you still get the same level of automatism and you still get the same benefit of speed, but it kind of comes from within. I think that part is probably something that would remain very valuable. So in general, I don't have anything against it. Again, I don't have anything against the SaaS category at all overall, but I have, as I said, different opinions of who actually is going to adopt to the future and adopt to the future and those who don't know. You mentioned this idea of people have a few extra hours over the weekend and they come in the morning and they're like, well, I had some extra time and I decided to do this. What's the coolest or most novel example of something that people basically vibe coded in a limited amount of time that wouldn't have happened say two years ago? So I've seen people doing cloud migrations of legacy applications to her on-premise once they have been enabled with those tools, a liter and a matter of hours. I've seen someone build a complete travel assistant for corporate travel assistant that looks at your calendar, it looks at the flight delays and look at the book and stuff literally like during a meeting where they were not paying attention. So those are some of the things. That's what I'm doing right now. While we're doing this podcast. That brings me to exactly what I wanted to go next, which is I'm curious like do large corporations have a token budget the way they would have a dollar budget in the past. So like I would love to have unlimited access to coding models and whatever and actually just play around and try to work on all that. It's all my favorite question. But I'm curious like how you think about token allocation within the firm and whether there's intro firm competition for compute. Yeah, token allocation could be like included in your performance. Yeah, really. If you do well, you got more tokens and stuff like that, whether that's part of what you think about for planning. Absolutely. So like a few months ago, I did that I spoke about predictions for 26 and one thing that I said was going to be the birth of the personal assistant and that kind of happened with open-close and all that stuff kind of early on. And then the one was this is going to be a token sticker shock for CFOs right that all of the sudden they're going to start seeing bills that they absolutely did not expect. Jensen Wong said an interview today or sorry not today in recent weeks. Something about like if I'm paying an engineer $500,000. I hope that he's spending at least $250,000 on tokens. Now again, as many people pointed out, that's like the barber saying, oh, you really need to get haircuts every week nonetheless. We're talking about some pretty big numbers a lot more than just like a cloud max plan for 200 right now. So talk about that right now. Okay. So first of all, lesson number one is you need to centralize the access to models so that you can monitor it and then optimize it. Okay. So the Wild West of everybody goes and calls an API and starts consuming tokens and then you find out later on is a big problem. And so that's why we built with this GSA platform, which is what's called a model gateway. And the model gateway intelligently routes requests to the combination, the Pareto frontier of quality and cost. Okay. So you got to centralize that. It's not a one size fits all because many cases if you're asking what's the weather, you don't need to call out the OPPOS 4.6. You can ask it to live in a local model that you're on very cheaply on-prem. And so there are ways to optimize the way before you start even having the conversation you're consuming too much. So this is very interesting to me. It's a big part of the problem that you're trying to solve. And we know like Chagy Bt, they intelligently route. They do some on there. You go to Chagy Bt.com and they'll try to route it to the best model. And there might even be some conflict of interest because they probably want to route it to the cheapest model. The user wants the most performant model. But how much of the work of your senior engineers is essentially solving this problem of the right query going to the Pareto optimal model? It is a big part of the time of the spent by the AI central group. Okay. The platform group. The platform group who is a lot about where do I get the right data for example for this question and which model do I write it to? That's a big, you know, because again, I spoke about Pareto frontier meaning the optimization between quality, which we don't want to compromise and the actual cost. And you can be ISO quality of very different price points because not all questions require the most expensive token. So that's point number one. So what I'm trying to say is my philosophy is to try to isolate the developer or the user from the token anxiety. It's a little bit like with electric cars. Okay. At one point if you have a team of range or always optimizing routes and maybe I'm not going to go there. I don't need this ice cream today. You're self limiting. You're self limiting in ways that are kind of really no useful optim, micro optimizations. We don't want people to go there yet at least. Right now it's a time where people need to really find the best way to kind of do more and to bet the best possible work with the AI. And let us, meaning internally in the sort of a central team, optimize it in a way that we're going to make it economical. And I think reducing the token anxiety is a big challenge, but I think it really frees up creativity and what you can do with the AI. It's also like there are certain problems that you don't want to optimize too early. Okay. Okay. So for example, yeah, how much time do you want to optimize now for a remember we used to kind of optimize the way to web pages because they were too slow to load. And then at one point the editor, so whatever say why can't I put yet another image and then people are starting to say, okay, why don't you do it. And then on the backhand, I'm going to work in optimizing your images rather than asking you what the most you can put three images on the homepage. Right. So that's the approach. And if people right now I would rather have them on the side of usage and let me worry about optimization. And the other point is really at the end, human hours always tend to be the most expensive cost. Okay. long as your token cost per hour is less than your wage per hour. That is a kind of a positive ROI. So at that point, it's fine. Well, what's your feeling about future costs of tokens and whether they're going up or down because you hear different things on this. One of the things you hear is that again, going back to the beginning of this conversation, AI has improved so quickly in the course of months, if not weeks, that those costs are destined to come down. But on the other hand, we know that the hyperscalers are still losing money hand over fist for power users such as yourself, at Goldman. So where do you think those are going over time? My personal view is token cost is going to go down quite a bit, but token numbers are going to go up. Yeah, probably more. And so total token cost is going to actually we're going to have to accept that it's going to be a major item of cost in any organization. And it's to be compared to the cost of people and not to be compared to the cost of the or TCP/IP packets or computer or any of that. If you look at just the number of tokens being used for the same use case, if you go the reasoning route or you don't go the reasoning route, if you go the agentic route or not the agentic route, if you go the open claw route where you know it checks every you know starts having these tasks that are firing one after the other and then you have to start to have verifiers etc etc. So I think the trend will continue with regards to more and more of those, but the cost the per unit cost of token, I'm pretty sure that is going to go down. Also because as GPUs are becoming more powerful, the cost per watt hopefully is going to go down. And then also like to be fair, I mean, these hyper scalers are doing a lot of optimizations to try to run those stacks on their own hard, right, which will potentially kind of also generate some economy. The news doesn't stop on the weekends. And now Bloomberg is the place to stay on top of it all. Hi, I'm David Gurra. Join us every Saturday and Sunday for the new Bloomberg this weekend. I'm Christina Raffini. We'll bring you the latest headlines in depth analysis and big interviews. All the stories that hit home on your days off. And I'm Lisa Mateo. Watch and listen to Bloomberg this weekend for thoughtful enlightening conversations about business, lifestyle, people and culture. On Saturday mornings, we put the past week's events into context, examining what happened in the markets and the world. That on Sundays, we speak with journalists, columnists and key political figures to prepare you for the week ahead. Join us as soon as you wake up and bring us with you wherever your weekend plans take you. Watch us on Bloomberg television. Listen on Bloomberg radio, stream the show live on the Bloomberg Business app or listen to the podcast. That's Bloomberg this weekend. Saturdays and Sundays starting at 7am Eastern. Make us part of your weekend routine on Bloomberg television, radio and wherever you get your podcasts. Can Goldman employees like run open claw on their work computers? And I'm curious like about the degree to which you have people who like want to, I want to install this or this seems like cool and then think about the security imperative and how you handle that aspect. Not the token anxiety, but the sort of, I want to install this. This is awesome. This is not my home. As you know, like as a bank, we're putting it locked down in terms of what you can install. You cannot install stuff that is not in the corporate, in the corporate app store in a way. And so there's no way. But you feel like a surbid. Definitely no way though, because you just ask Claude how to install itself. But he's not going to be able to execute. He's not going to be able to create the actual executable. He can actually even GSA assist them today can spit out a lot of code, but it's without source code. Now what source code doesn't run executable. So it needs to be built and it needs to be turned into an executable. He needs to be signed. Otherwise, the operating system is going to refuse to run it. And so it just doesn't run unless you have that. But do you feel an anxiety where startups, they're probably and there's not a sort of startup investment banks, but there are various spin-tax and other things that want to chip away at parts of your business. And they can run perhaps faster and they can be a little bit more liberal about what their employees are allowed to do, etc. Do you feel like you have to keep a certain cadence of expanding the list of those executables that are able to be run? So I'll give you two answers. So first, I want to make sure that I answered your first question, which is we're not using OpenClaw. Okay. Okay. But some of the properties of OpenClaw has actually informed the way we are building our agent platform. Okay. Agents today, because of OpenClaw, have actually changed. If you break down what OpenClaw is, there are it is my own interpretation. And I actually never even spoke about this. There are three characteristics that make OpenClaw what it is. One is it's a constant loop. So it's basically what in information theory, you can call an observer pattern. It's something that continues to run and observe. So there is that. It's a constant observe. So runs constantly. The other one is it can schedule events every 7 am, do this or I, you know, like in personal life, we all have something like that that sends me the news in the morning and all that. So there is a schedule ability of tasks. The third one is you can instruct it to kind of change its own behavior. Because it has these files, dot MD files, soul, dot MD, the way you, so you can say things like, hey, I would like you to never use this term or please change or change the way you filter news. And so it kind of writes its own software to do things for you without you even seeing what's behind the scenes. And so instead of letting people install OpenClaw on their computers, what we do is we incorporate some of those characteristics into our agentic platform so that it does things that are more similar to OpenClaw. So that makes a lot of sense. Your second question was interesting because basically if I read behind the questions, are you asking whether there is a sort of a velocity disadvantage with regards to us versus others? I often say that there is a difference between speed and velocity. Speed is almost like you have a certain sprint, okay? But then at some point you're going to hit the wall, security wall, scalability wall, there's going to be a bug, you don't know what you're doing and it's going to first sooner or later you're going to be hit by that. It would be like most airplanes are in autopilot that can do everything. So theoretically you and I could go on the cockpit and for a long time during the flight, we will feel pretty good about that, right? We will drink some soft drinks or your tea, we will maybe watch some videos, we will be very happy. That sounds great. So there is one point at one point in time in the flight, there's going to be some storm and there's going to be an autopilot disconnect and you and I are going to look at each other they're going to say, oh, right? Where's the pilot? Could I just say recently I was on a flight to tell you this that I was going to Newark? Oh yeah. We circled three times, we tried to land, it was during a storm, they kept not landing. Everyone was like starting to get pretty annoyed because we were up there for a while and then the flight attendant comes on and says, unprompted, by the way we have plenty of gas and everyone started, we got, no, it's like this is a question, this is an answer that no one had been questioned. Nobody need anyway, I'm sorry, I just want to get there. Then everyone got really nervous. Okay, by the way we have plenty of gas. But you know what, I was almost fearing that you would say is there a pilot on the wall? No, no, it wasn't. And then we did land in Washington DC. Okay, oh you did. Yeah, from Newark. No, no, that's not good. But that makes sense, yeah. And so and so that's why I mean by velocities really like is like the marathon is a sustained speed for a long time in a certain direction. Yeah. I don't think by randomizing that you actually gain velocity, you gain an instant speed of some sort. And so I'm kind of optimizing for the loss. But related to Joe's point though, you are a regulated bank. Absolutely. Right. And so there are restrictions on what you can do in terms of technology. I am very, very curious what your discussions with regulators are right now because a lot of regulators, this is still pretty new to them. A lot of the models basically are black boxes. How do you convince them that like they're running as they should that they're spitting out the correct output that you understand how they're actually functioning? So this is not the first time that banks use neural networks. Okay, these are just much larger neural networks. But we've been using neural network for like a decade plus. And so every bank has already gone through the motions of explaining that neural networks don't have perfect explicability. Therefore, you need to change the control system around them. You need to look at what actions can they actually do and then you limit the actions. Okay, there is functions called the like a model risk management, which is a very standardized you know, function within every bank. Therefore each of those neural networks, you need to have an inventory. You need to have a risk tiering and you need to put controls around them. So it is not really that much of a new thing is more of an evolution where now you have things that are much faster and much more powerful. But the basic pattern and the basic discussion with regulators is kind of the same, which is are you classifying the risk tiering of the application, right? And then which controls are you putting and are you putting human supervision and human in the loop? So for example, for code, we don't allow AI's to auto-approved the wrong code. Okay. All they can do is publish what's called the pull request or a merge request. The same way as a developer would do. And we kind of have a sort of a zero trust model there, because we don't assume that maybe a junior developer is going to be less more bug-free than an AI, right? And so we have several controls in place, for example, there needs to be a human, it's more senior than you that actually looks at the code and then certifies and approves. And then after that, before it goes to the production, it goes something that is called CI/CD or continuous integration, continuous deployment, pipeline, where when it goes through the build phase, etc. There is a lot of checks that are injected into that. There are security checks, there are tech risk checks. So I don't think at the end of the day you really lose too much velocity or at all. You just need to invest more in those kind of things. And the regulators, I think, if you bring them back into sort of a familiar territory and you're also honest on things that you know and things that you don't know. And for the things that you don't know, you kind of put higher protections. I think the conversation is generally very positive. We were having an episode that we recorded several weeks ago that we still haven't released. I don't know exactly the timing of that one of this one. We interviewed Scott Bach, the former CEO of Green Hill, the boutique investment bank. And part of the reason we had that conversation was because we want to know, like, if AI is going to someday disrupt banking as we know it, we're talking about the history of investment banking. But one of the things that he talked about was that a big advantage that the banks had was this sort of information asymmetry and that they would know a lot more about their industries and so forth than their clients. And this was profitable. Now going back to your answer, the very first question, you're like, okay, a client might call gold men and they say, what is the straight of foremoose closure mean for this portfolio shock, etc. I was going to ask this exact question. I kind of think I could do that. I think I could, I mean, no offense, I'm sure your platform is a little bit better than what it, like, but I think I could like get 90% of the way there. And I bet I could like with a little bit of data build a basket that says I want helium shortage basket, which country, which companies would I short if I think the helium shortage is going to get worse? I could buy build a basket the way a trading desk would. Do you think long term like that AI roads a certain structural source of profitability for banks and you were about that. I think you can get to the 90% but I think clients are really paying us for that extra 10%. Okay. So I think that's the answer. So what is the extra 10% in that context? Is it the models are slightly better or is it also the data? We have access to, you know, we buy a lot of data that is very expensive and it's at massive quantities and it's very up to date and very real time. So we have a little bit of a data advantage. We operate across multiple asset classes. So we see the trading side. We see the asset management side. So we have a sort of a correlation between assets advantage that we see those because generally rates move, interest rates can move yields can move. You know, there is a correlation between all those indicators. So there is another advantage. We have a global advantage. We have people on the ground in 100 plus countries and these people have relationship and information traversed through those channels. And also, you know, like we generally deal with very complex portfolio. So this is not you and I maybe having three stocks or four or five. This is like very complex multi assets with complex products like swaps or swapsions or exotic products, etc. And so that's really the 10% that the clients that we have really value and what we really need to get. It's like at the end of the day listen. Look at Formula One. The difference per time, per lap between the Mercedes and take, you know, your favorite last team. It's sometimes one second after two minutes. And that is the difference between, you know, getting a $100 million a year sponsorship or a $10,000 sponsorship. So for sophisticated clients, that 10% is really what the money is. And that's really what people are paying us for. So actually you mentioned all the different businesses at Goldman and there are a bunch of them like asset management. There's banking, there's trading. A lot of those businesses aren't supposed to talk to each other in various ways. And so when it comes to the data, is there like a data leakage issue where you might have a model that's in house like GSA. That's pulling data from different sides of the company in ways that maybe it shouldn't be maybe it's really hard to tell given the complexity of the model. Is that something you have to pay attention to? Absolutely. So we have the concept of info barriers. And the info barriers are enforced throughout the entire system. And they're linked to your idea or your account. Okay, so if I'm on the private side, I can only see certain information. If I am on a public side, I can only see certain information. And I cannot even know about the information on the other side. I can't, I don't have access to the files, to the folders, nothing. Each AI or each agent or each application, that's the beauty of this centralized platform, needs to get an idea or a badge. And that badge is attached to the exact same info barriers as any application or any computers. So this is an enforced basically at the source. So even if it is the same type of model, but the particular use of the model, the particular session of the model, they need to get a ticket or a badge. And that badge or those keys just take them to certain place. And so this one or this being, it took us almost two years to build a GSA platform. This back to the reason why you can't be casual about these things. This thing has not been built by some random vibe coders. Because you need to worry about cyber, you need to worry about info barriers, you need to worry about all that. And so when I talk about, there are places where you can leverage and do correlations, but there are others where you absolutely can't. And this is kind of, that is foundational to the fact that you need to be ready for AI. You can't be casual about AI. So I take your point that there's never been a technology that you've seen in your career that has actually reduced the need for software engineers. And that the nature of the job of software engineers is a change, maybe gets more high level and whatever. Setting aside that volume question, setting aside the pure head level of head count question. Is AI changing right now across anything technology or otherwise the types of person you're looking for or changing something about the nature of the type of talent you're person. Yeah, absolutely. Great question. So I think in this day and age, almost nobody is an individual contributor, really. Because when you're working with agents, you need to have at least three fundamental characteristics. One is you need to be able to explain what you want to get that. The second one is you need to be able to delegate work. Yes, what? Because you're going to have multiple agents. One is specialized, for example, in doing, I don't know, DCF calculations. And one is specialized in doing research. So you need to be able to break down the work into chunks that can be executed in parallel in some way. And then three, you need to have the ability to supervise. You need to actually look at the output and say, okay, I'm good with this or go back. It turns out that those three things I explain, delegate and supervise are kind of the one or one of managers. Managers need to have those three. Otherwise they can't manage a team. And so AI is kind of turning everybody a little bit into a manager. And those are kind of the skills that we are actually looking for people that they know that they're going to have agency on tools that at some point are going to be even more proficient and specialized than they are. And so the most important thing is really the ability to ideate, to explain, to delegate and then to really know what good looks like. And I think that is a big change. And I don't think everybody is going to actually rapidly go through that. And I think we're doing a combination of training. There is a combination of exposing them to other people. One of the advantages of having forward deployment engineers is also that there is a little bit of clash of culture that is happening on the table. And so people think really, really differently. And that pushes people outside their comfort zone. That's why I'm saying that there is a little bit of a metamorphosis happening there is not just about the fictions is really thinking about is my job going to stay the same. No, it's actually changing quite a bit. So I'm thinking how to frame this question. But what's work life balance like now for a developer at Goldman because you have this existential angst about jobs potentially changing. At the same time, you have AI tools that enable more productivity. And you also have this thing happening where I feel like Joe maybe no more about this than I do, but I feel like a lot of vibe coders like like it's addictive. Yeah, right. It's like you're pressing the button of a slot machine. You're interacting with Claude and you're seeing what it spits back out over and over again until you get that big win. And so I've heard people talk about burnout among developers who are just doing so much with this right now that they're just hitting that button over. There was a good discussion in the AdLod discord recently about exactly this some engineers and semiconductors feeling that the job has become less satisfying. And I think it's sort of what you're getting at. There's sort of slot machines. Yeah. Where it's like, oh, you're getting like hit the prong. Okay, this is the great output. Then it's like they feel the workers like less satisfying and stuff like that. Then actually like writing code. Yeah, I mean, listen again, this work kind of the fact that I'm a little bit older than most here engineers kind of I've seen that the first time people had the Excel. I've said for the first time people had Python. Oh my god, I don't need to know Java and then the kids start to code and there is this whole coding movement and then you get to start creating your applications. I've seen the front. first time people have mobile stuff and mobile apps. And so I think a little bit of that is because it's new to be perfectly honest. And I think, yes, there is a little bit of that, but there is a little, a lot of novelty to that. And then I've seen that people have been using those tools for a couple of years. They're taking them a little bit more like, okay, it's professionalist tool and I'm gonna use it for what I actually need rather than just trying to discover. One thing that I've seen is that because maybe of that, but also because of what you can get, there is a sort of a, in a way, reward cycle that is pretty quick. Yeah. People are very excited actually. There's some sort of a joy of the profession that is actually coming out as if engineers were feeling like this job is new again because a lot of engineers have seen the same patterns sometimes for three decades. So that has been something that I observed. There is also a lot of peer pressure. There is a lot of fear of missing out. So people are rather than, is no longer me trying to push the car up here. More people are actually looking at their peers and they're looking at, oh my God, how could you do that? And so it's kind of spreading horizontally quite a bit, which is really nice to see. And so, so far I have to say that it's been positive, a positive change and also one other thing that we're talking about burnout. I see that a lot of people get fatigued. I don't wanna talk about burnout, but they get fatigued when there are a lot of repetitive tasks, especially for a developer. Here's an example. Let's say you go from a version of a Java library or Spring Boot to another version. And then all of a sudden you compile and you get, or you build and you get all these errors that says you need to upgrade. Honestly, upgrading libraries is not the most fun job. And if you need to do it a hundred times, or it's like someone says, by the way guys, we have this new design, new logo, new colors, implement it on like 200 websites. It might be fun the first 10 and then it becomes a drag. And so I think taking that away, kind of they focus more and more on the plan, for example. And so right now let's do a migration plan to the cloud of a complex application. They spend maybe 70% of their time going back and forth with a very powerful set of AI's to really get the plan right. They feel a little bit more elevated. And the mechanical part, it's kind of left to the machine. The same way, I mean, listen, I started developing when I was literally flipping switches. And then pressing a button, which will move the register up one. And then came some languages they like see, oh my God. Now I don't have to flip switches anymore, but guess what? I need to do memory management. I need to do pointers. I mean, there's a lot of heavy lifting. Oh, I have a memory leak. I'm gonna spend a week before I actually finally identify that. And then it comes Java or garbage collection. I don't have to worry about memory leaks anymore. Fantastic. And then comes Python, which is all that rigidity all so much easier to be type free and so forth. And so every time you kind of keep raising the bar and a lot of the kind of mechanics kind of goes away, I think this is being like a 10 years jump in a metro to Ears, but I think overall, nobody really likes to have that toil and that mechanical work. And I'm actually quite happy that people are gonna spend maybe initially more time because they're excited, but don't think that they're enjoying rather than think that they just they'd read. All right, well, Mark, we'll have to have you back on the podcast in another year and a half, I guess, and discuss, yeah, with the world of change again. On the reduced AI timeline. Thank you so much for coming back on the podcast. Thanks for having me. Thank you so much, Mark. Thank you. (upbeat music) - So Joe, that was great to catch up. One thing I thought was really interesting was his point about the discussions with the regulators and framing it like very similar to previous technological advances where you're not necessarily explaining exactly how the models are coming to certain conclusions. But you're more focused on actually limiting the risks and making sure that they're in the right bucket for risk assessment. - No, I thought that was really interesting just that some of these technologies, the black box. - Yeah. - LLMs are not the first black box. - I mean, we've actually been talking about black box trading for years and finance before. So the idea of like, okay, there are these things that are happening, we can articulate them and whatever, like it's not the first rodeo for finance. It's really interesting. And also, I thought the whole conversation about token budgets and allocations are interesting. The idea of like, okay, part of the job here is you have a bunch of different models. Everyone in theory wants the most performant model, but how do you find that optimization where you get the best performance relative to price? It sounds like a pretty interesting, like engineering problem. - Yeah, I would actually love to do more on that question. - I would too, yeah. - Because it's such an interesting question of incentives, right? And like, how do you actually like prioritize which projects? - And how do you know what constitutes a good output? And how do you, you sacrifice a little bit of quality for like 10X less token budget or whatever? Like, there should be, it would be very interesting to talk about how that problem specifically gets solved in the type of an organization. - Token economics or efficiency optimization. Yeah, well, we'll have Marco back on very soon to talk about all the new things that AI is doing. But for now, shall we leave it there? - Let's leave it there. - This has been another episode of the AdLots Podcast. I'm Tracy Alleyway. You can follow me at Tracy Alleyway. - And I'm Jill Weisenthal. You can follow me at the stalwart. Follow our producers, Carmen Rodriguez, @CermanArmenDash, she'll be in it at Dashbot and Kale Brooks. And for more AdLots content, go to Bloomberg.com/AdLots or the daily newsletter on all of our episodes. And you can chat about all these topics 24/7 in our Discord with fellow listeners, discord.gg/AdLots. - And if you enjoy AdLots, if you like it, when we talk to Goldman Sachs about how they're actually deploying AI across the company, then please leave us a positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely add free. All you need to do is find the Bloomberg channel on Apple podcasts and follow the instructions there. Thanks for listening. (upbeat music) (upbeat music) - I'm Francine Lacqua, an award-winning journalist. And I've got a new podcast. Leaders was Francine Lacqua from Bloomberg Podcasts. I've interviewed everyone from heads of state to fashion icons about the news of the moment, but I've always been curious who are these people as leaders. I don't think there's one right way to be a leader. - Make decisions. A poor decision is always better than no decision. - Listen to new episodes every other Monday. Follow leaders with Francine Lacqua, wherever you get your podcasts.