Video: AI is changing jobs | Duration: 2880s | Summary: AI is changing jobs | Chapters: Welcome and Introduction (2.32s), Speaker Introductions (90.66s), Meet Kathleen Herrera (161.955s), Fluency vs Capability (288.41s), Human Elevation (641.75s), Technology and People Collaboration (692.69s), Culture Over Tools (1050.55s), Shadow AI Management (1210.435s), Human Judgment and Accountability (1467.805s), Human-AI Collaboration (1643.81s), Redesigning Work with AI (1899.935s), Redesigning Work Workflows (2064.075s), Measuring AI Success (2205.675s), Rethinking Performance Management (2367.18s), Upcoming Event Invitation (2467.525s), Q&A Session (2571.685s), Executive Alignment Strategy (2633.065s), Closing and Trust (2754.78s)
Transcript for "AI is changing jobs":
Hello, everyone. Welcome, and thank you for joining us on this lovely Thursday. While the rest of us are still coming in, I'll start with a few housekeeping items. Firstly, please share your comments and questions in the panel on the right hand side. Our speaker speakers will dedicate the final ten minutes of the webinar to answering as many as possible. And, yes, for anyone wondering, today's webinar is being recorded, and the replay will be sent directly to your inbox afterward. While we wait, we'd also love to hear you from in the chat. So please tell us where you're joining in from, what is your role, and how, if at all, has AI changed your role over the past few years. It's a simple question, but it gets to the heart of today's conversation, And that is that AI has changed the work people do, and it doesn't just change the task at hand. It changes how roles are structured, how managers lead, how decisions get made, and, ultimately, who is accountable. And so that brings us to the bigger question. When AI changes the work, what else has to change around it? And to help us explore that, I'm joined by our experts, Dan and Catalina. Dan, without further ado, I'll hand it over to you for a quick introduction, and then we'll hear from Catalina. Brilliant. Thank you, Renata, and it is such a pleasure to be here today with Catalina. And thank you to all of you who joined us. This is a very timely, very important topic. I'm Dan Turchin, CEO of PeopleRan, the AI platform that automates employee service. We're out to give the next billion employees back about four to six hours of productive time per week. We deliver what we call as human first AI. We're gonna talk a lot about what it means to be a human in an age where we're increasingly asked to collaborate with machines. On the side, I host a podcast. It's called AI and the Future of Work. We've published over 450 conversations. We have a community of more than a million listeners. One of our great recent guests was Florian, who's the CEO of Dataiku. We became friends. It's a great conversation. If you haven't yet heard it, maybe we can share a link to it in, in the notes. And, after reading the Dataiku report on this topic, I was very enthusiastic about joining my new friend Catalina and, participating in today's conversation. So looking forward to this one. Catalina, over to you. Yeah. Thanks, darling. Phenomenal intro, and I I agree with you. This is super exciting times. I'm Kathleen Herrera. I am Field Chief Data Officer at Dedeku. What that means is that I spend my weeks, inside AI programs, the ones that are going very well, also the ones that are going a little bit sideways. I'm a very human, and I like to say that I'm a human empowered by AI because, in addition to my field chief data officer role, I am a DJ. I compose my own music. I'm using AI to compose my music. I mix my own music. I also am the president of the Southeast Textiles chapter of High-tech Rising, which is a nonprofit organization where we, target the esteem, closing the gap on esteem, and opening doors, and building mentorship and giving girls hands on experience and identifying the opportunities within tech. I also co own a paint by number company, so I am very into art and music and creativity. So in my right side of the brain, I am a data scientist, and I have been a data scientist for the last twenty years. And I have seen the change on on every chapter that we go through. Right? The Internet, what do we do with that now? All the components that AI brings to the table. Again, how are we, the humans, empowered by it? I'm very excited about this conversation today, so thanks, for having me here, Diane. And, Catalina, I should mention that even though Renata introduced you and I as the experts, I'm gonna just lead out by saying everyone in the audience that has chosen to join us today, you're the experts. So please take us up in the offer, engage in the chat, or offline. I'll certainly be happy to share my contact information. These are conversations that can't be, confined to to one short session with, with a handful of full of questions. So you're the experts. Let us know what you think as well. Catalina, let's get started by a fairly deep question. We're gonna jump right in a Forbes article. Found that confidence with AI doesn't always translate into real capabilities. We're living it in real time. As AI reshapes job functions, when does tool fluency stop being enough? And what capabilities become more important? Maybe you lead out, and then I'll, I'll reciprocate. Yeah. No. I I love this question. Right? It's it's kind of questioning what is it that we are measuring here and how do we focus this? And it actually touches on something I see every single day in the enterprise where we confused fluency with capability. I will say that to fluency, I know how to write a prompt, I know where to use the user interface to write the prompt. That is just the price of entry. I think about this like knowing how to type. The fact that you know how to type is not going to make you a great writer. Fluency is starting to be short when we think beyond that. When we see all of these AI outputs being generated so fast without us questioning it, then we have a problem. That is when we start thinking about what else needs to be part of this equation because this is a multivariable model. Now it's not only the tech for the sake of tech. We know that the intelligence layer is now there, that we can pretty much rent these models as part of the intelligence layer that we are using as part of our daily basis, but the real capability comes down to, I would say, three things. The number one is the judgment. We, as the humans, anyone can generate a fast report or whatever, but the value comes from questioning the output. Is this accurate? Is this considering everything that needs to be considered? Is the context missing for the model? You know, that's number one. Number two is intentional accountability. Now we know that we are using these AI components as part of this mix, but that doesn't mean that we are removing the responsibility into it. I think actually the opposite. It redistributes the responsibility. And I will say that as a leader, we have to own now more than ever the ethics behind it, the trade offs behind it, the actual measurable impact behind the data. No? And the third one that I think is super important is that human orchestration. So noise is not only automating execution on a micro automation job. No. It's actually the adaptability, the trust, super important, the trust, and the coaching. Those become your most critical hard skills, no? Because the the tool fluency is gonna give you speed, but the human judgment gives you the direction. And when you are scaling AI, I mean, garbage in garbage out, and AI is gonna scale whatever we give it. So our job is to make sure that we are elevating the human intent and not just automating tasks. How do you see that, Dan? Very well said, Catalina. I'll add to that. I frequently coach teams to replace the term AI with math. Whenever we're at a juncture where we feel intimidated by the emergent properties of AI, it's easy to ground ourself in the difference between what a human does and what a tool does If we think about math, it's a skill. It's a competency. It helps us do a job. It helps us maybe even explore the world, but it doesn't have human intelligence. It doesn't bleed. It doesn't have emotions. It's not sentient, although I know that's a debate that many like to engage in. And so when we think about, what it is that makes us uniquely human, it's important to also acknowledge that this is a different period that we're undergoing than previous technological transformations going back to the eighteenth century and the first industrial revolution. Because Mhmm. Yep. when we introduce, let's say, the steam powered loom that let us, weave clothes faster, it wasn't quite like AI where it's, almost like holding up a mirror to our own humanity. It didn't speak to us, and it didn't feel human like AI does. Nonetheless, we always have to remember that when we see ourselves through that mirror, it is the case that AI only knows what we tell it to know, and AI only does what we tell it to do. We have the agency. We may choose to assign agency to AI. But whenever we get into these complicated arguments about it feels human or it feels like it is agency because of some of the things that it appears to do intelligently, let's just always remember that, ultimately, we choose how and when to engage AI. And it's okay. Let it shape our jobs. We'll talk a little bit more about what that means in subsequent questions, but, let's go in with a learner's mindset, and let's not be intimidated by it. Let's be willing to embrace what it can mean to have a machine maybe complement some of the, some of the capabilities that humans inherently lack. Exactly. It's. not about human replacement. It's about human elevation. And and I love your steam analogy. Right? When we invented the steam engine, we didn't stop meeting people. We stopped asking people to provide the power themselves and started asking them to design, operate, build, and imagine an entirely different scale. Right? Same with the calculator. I like your math analogy as well. Right? Not because we have a calculator, we don't we don't do math anymore. No. The problem is I'm getting more sophisticated. You said two important things, We see three. direction. If we always come back to. those as. the as the themes, then then everybody ends up on the right side of innovation. Kind of a natural way, Yeah. Exactly. to, kinda segue into second topic. What becomes possible when technology, data AI, and people design AI adoption together? I know through the great study that Dataiku published, we learned a lot about some interesting, case studies. What gets missed when they don't collaborate? Yeah. And I think this is an important one, and I think this one comes from twenty three years of of of learning. And, Yes, we've done that and learned that. I think the summary of the situation is that the technology keeps moving. We are talking about the last twenty, thirty years, big evolution. Think about the first USB that you pay, the capacity, the last USB that you paid for, the capacity, and you see that exponential growth and everything that is compute and hardware. That is going to continue to change. I will say that opportunity that we have today is to acknowledge the power of that technology piece and component, but also acknowledge that it's just one of multiple variables that need to be considered for this to be successful. And when this when we say successful, what it means is that we are empowering our domain experts, our humans that know about this specific business use case or that know about the question that they want to ask the AI, but we are empower our people. No? So when tech, data, and people leaders design adoption together, you stop doing a so far rollout and start engineering the true organizational capability. We have a multivariable model that includes the culture. When it comes from a holistic perspective, you have a vision. What we say is from the top down. As an enterprise, we know what we want to achieve. We know that we want to empower our SMEs with this technology, but we also know that it's not only the people, it's the orchestration, it's the governance. So we have to be very aware of the culture and how, we empower them, the employee to feel supported with a framework. So if the culture breaks, the afraid employee either is gonna ignore the tech or is gonna use it passively just to look busy. So governance. If governance fails, the tools cannot automate workflows, but they can't automate our own ethical trade offs and the real world accountability. See, if a manager gets stuck, the middle manager is going to be forced to drive performance without the coaching tools to reshape their team. So we need to think about this as a multivariable model that comes from our initiative of empowering our humans to to be elevated. So let's use it. And be the people, leaders, and the context, and why are we doing this, and why is that star that we are following. Right? But it has to be with that proper framework so the people feel supported. We have a we have a north here. We are aligned on what we are trying to accomplish. No? Absolutely. I frequently say that there's never been a better time to study the liberal arts. Now's a great time to be investing in learning philosophy or history. Read the classics, because when you enter the workplace, those are the skills that will distinguish you from what AI can do. One of the things that I frequently say is that the pace of technology innovation vastly exceeds the pace of organizational change. And what I. mean by that. is that, humans are complicated, and the the nature of what it means to deliver great outcomes as a company or as a team, in any in in in, in any industry, involves people to collaborate. and communicate well. There's recently a study conducted with about a thousand enterprise organizations by Microsoft and McKinsey, and I found these particular findings to be noteworthy, relevant to this this topic. 75% of employees, this was done recently in 2026, say that they use AI tools in their everyday jobs, and yet, 65%, only slightly less, say, they don't feel empowered to use those tools because organizationally, whether it's comp structure, incentives, processes have not been modified to take into account what the AI can do to empower the humans. Mhmm. And so as a result, Mhmm. there's this growing trust gap. We've all lived it. In fact, one of the other findings from that study is that there's a 30 percentage point difference between those organizations where employees say they feel supported and empowered to use AI versus those that say they're not. So it turns out, you know, this question, could probably be asked in even, you know, more, provocative in a more provocative way is the single. biggest thing that determines organizational success with AI, whether or not you're prepared to, keep up with the pace of. technology change by rethinking what it means to deliver outcomes and and, you know, kind of the the, the throbbing heartbeat of the organization involves a whole lot more than technology. Setting is a very important topic. Yeah. And and and second that, that if you design AI adoption in silos, you get expensive software and shadow workflows. Right? But when tech, data, people, leaders build together, you create a resilient human centered culture where technology amplifies that human potential rather than attempting to replace it. So is what is the framework that is gonna allow us to look at this in a more holistic way, understanding the different pieces and the different components, but also what makes you different from your competitor that has access to the same AI commodity intelligence. And that is the cultural part that becomes so important. Right? Because it's not only, oh, yeah. We are all using models. We are all super intelligent. Well, yeah. How you are using it differently and how that correlates with your culture, your processes, the way that your teams work together so you bring that differentiator to the table as well. No? So I I like the analogy between the Netflix and the blockbuster. How do you ensure that you are the Netflix and not only assume the change and empower your domain experts to adjust to the new piece of technology, streaming, for example, versus a blockbuster that never adopted that streaming capabilities and they went out of the market. Right? So is is this is the opportunity. This is the moment, and and this is happening as well. So how do we not only embrace it, but also ensure that our workforce is as empowered as they can be? All said. Encourage leaders to appreciate. I know this seems like captain obvious speaking, but, the fact that when you write a check to OpenAI or to Anthropic, you get a tool. But when you invest in the humanness of work, when you create a psychologically safe place for employees, that's when you can do. transformative things. And that kind of. equation, that way of thinking about technology transcends AI and every other technology, even what will come in the future. Something I frequently tell my kids, which I think applies here, nobody cares how much you know until they know how much you care. And I think we need to remind ourselves of the importance of that when we're having these discussions about, you know, what we think AI might be able to do, just because we write a check-in, and and onboard a tool. 100%. 100%. So I'm referred to that great, piece of research released by Dataiku, Davidiku Harris poll. Found fifty four four percent of CIOs discovered employees using unsanctioned AI. I know this is growing red meat to, Catalina and myself, the lions. How can leaders learn from shadow AI without creating an AI police culture? You go first, Catalina. I'll sit up my hands while you give your answer. I I do. I I love this one and I see it so so much. I mean, it's it's very tangible in the field, but my 2¢, I mean, Shadow AI is not rebellion, it's organizational feedback. Something is failing. It's telling you that your official tools have friction and your people are trying to solve real problems. You have two options here. Back to my Netflix versus blockbuster analogy. So you can ignore it and and become the if you play the I Police, and that is not gonna stop the behavior. You are just gonna drive innovation on the ground and take people to hide it. And that is kind of contradictory. So, my 2¢ is instead of becoming the eye police, you have to shift to curiosity. So let me ask you, what is the problem that you were trying to solve instead of who broke the rule? High Shadow AI is is actually a sign of high employee engagement because they are trying to solve the problem with the technology that is available for everybody else. Yes or yes. Because not inside enterprise, outside enterprise. We have access to all of these, like, levels of intelligence. So it's gonna be very hard for somebody that has access to all of these outside enterprise to be forced to old or friction, full of friction processes that don't let you be and and maximize that opportunity. Yeah. So shift to curiosity, ask what the problem is that you are trying to solve, and you have to pave the desire path. It's like when we see people walking in the street, crossing the grass, what is a smart planner gonna do? It's gonna make it a road. It's gonna pave that shortcut. Right? So don't build the fences, pave their path with secure enterprise grade guardrails. So you know that the trust is gonna be embedded because you have all the guardrails in place to maintain that human judgment as part of that flow. So it's not only AI given an AI output, you have those guardrails as part of that framework that is empowering your your your personas. Right? And you have to enable. Don't gate kid. The last thing that you want to do is become the bottleneck here. You need to make governance about safety and quality and not about stopping work because this is going to happen yes or yes. So the faster you, number one, acknowledge that this is a multivariable problem That is not only about the technology and which model are you gonna use, but it's also how you're gonna empower your people, how you're gonna follow your processes, and how you're gonna enhance your culture. Right? And then stop saying no because what that is going to do is that you're going to have more and more churn. I mean, it's going to be a workforce that doesn't want to work with a company that is full of friction and is not empowering their domain experts with the technology that is available in the market. No? So it's also an moment. So Shadow AI proves that your people want to be more productive and capable. And our job is not to police their ambition, it's to give them a safe, trusted environment to realize it. Those are my 2¢ there. Thanks, Catalina. I mentioned earlier that, you who chose to join us for today's session, you're the experts. And please, if you're not, do read through the comments. I see one from Slobodan. I'm I'm skimming them while while, Catalina and I are having a conversation with so many great insights. I'll add a bit to what Catalina shared. At the point where internally, you're having the conversation about shadow AI, once that thing gets named, you've already missed the narrative. Our job as employers, as leaders, is to create safe places where employees can do the best work of their life. And anytime an employee feels like they may or are being penalized for their usage of AI, Something has already broken in the culture. And so our job should be to create the right guardrails, provide the right way to say these are the tools that we support, but you go off and experiment. Because you know what? We are all in this together. And the questions that your team members have, your organization has about what we should do and how we should do and where it might add value, nobody has the answers, and that's okay. Now is the time for us to just celebrate experimentation. That's another thing that separates us from the bots. We have hypotheses. We have ideas. We can pursue them. We're not always right. But I guarantee you that together, when we tap into the collective wisdom and the creativity and the inspiration that comes from all four corners of the organization, we're gonna quickly stop having a conversation about AI police. And instead, we're gonna create that environment where we can outperform the competition, but mostly, where employees feel super engaged and trusted and valued and respected. And, you know, back to that Microsoft McKinsey study, like, those are the organizations that are doing the best. Those are where employees come into work, and they've got a little spring in their step. And they say, now is an awesome time. I am doing the best work of my life because I have this thought partner at work, because I have access to intelligence as a service. And that is just the. antithesis of any organization where you talk about, you know, the need to police AI or, Mhmm. or mitigate the potential downside of, quote, shadow AI. So, yeah, please, you know, take that into account. Keep the feedback coming. To the extent it's it's, Uh-huh. possible, we'll we'll augment our comments with your comments from the, from the, chat panel as well. Catalina, let's move on to topic four. As AI becomes embedded in more workflows, how should organizations think about where human judgment, accountability, and expertise add the most value, and how should roles evolve based on number one? Yeah. I will say this is an important question as well. And I will say that we have to be human value moves to the bookends of every process. So it has to be at the beginning because you need to define the right problem and define the strategy. AI is going to give you the answers, but humans are going to bring the intent. That's the beginning. Then at the end, where you have to add your critical judgment and accountability. So is this accurate? Is this within these cartels? Is this spelling my transparency test, my bias test, my drift test, my data quality test, everything else that I have to put in place for me to feel that this is an accountable AI outcome. No? So AI offers probabilities, not truth. And and especially now that we are renting this intelligence layer, I am talking about the LLMs and the generative AI layer that is part of the ecosystem now. So we are talking about non deterministic answers that can offer a lot of probabilities, but not necessarily the truth of the truth within your business and your business use case. So, humans must audit the logic and own the outcome. So, I am the domain expert, I am the one who is doing the best, the what if scenarios, the is this an hallucination or is not? Is this being compared with a baseline or bank of true positives from history? Like, we are the ones that know them context and understand what needs to happen. So because of this, the roles are shifting. And and they are shifting in, I would say what, three big ways. The first one is that the operators become orchestrators. Now you have artifacts through history and through technology. So you have the infrastructure from the hardware aspect. You have those silos of data information, then you have data artifacts that come out of those silos of infrastructure where you are transforming that data into predictive, into, generative, into agentic, into AI. Right? So you need to now orchestrate, you need to stop playing every instrument and start conducting the system. No? That's your human role now. Now, second, design ops. The sign offs are gonna become a kind of an active stewardship. You're driving that. You don't just sign off because the computer gave you a need pretty report. No. AI computers can generate reports that look completely correct on the outside, generate reports that look completely correct on the outside, but are showing you the wrong information. Example, I was using a retrieval augmented generation chatbot that was using all the policies that were not updated. Or you use a GPS that is not using the most updated maps. You may get follow the directions, but the direction is not gonna take you to your final destination because it's not even part of the information that the AI is analyzing. So these are the very important things that we have to consider here. And then the third one is that the managers are becoming coaches. Now it's like, okay. With with, the opportunity that we have here, we can stop being the traffic cops and start coaching critical thinking and trust. Where is your human insight here? Is this an accurate outcome? Can be customer facing? Do you have a bank of actual true answers historically given? No? So AI doesn't push human qualities to the margin. It actually brings judgment and empathy and ethics directly to the center. And that's where I would say that we are right now. Super important role more than Thank you, Catalina. Bob's comment, I think Bob Federici's comment in chat says it all. We leverage AI to help evaluate a project on paper. ever. It all look great, but it turns out, you know, without applying before you applied human thought and experience, you know, what we. thought looked good on paper just kind of proposed by AI really wasn't gonna work. And so it's that combination, what I call, you to the power of AI that delivers the best outcomes. I'll supplement that with, a framework that PricewaterhouseCoopers recently published, very relevant here talking about how goals and careers and jobs are evolving. They divide work into two categories, those jobs that can be democratized with AI and those jobs that can be what they call professionalized with AI. I think it's an important distinction. Mhmm. First off, they know. that all jobs, obviously, will be impacted by AI. What they mean by democratization is AI allows the person in that role to do more of the old work faster, versus professionalizing a role or a job with AI. What it means to professionalize that role is to figure out the right work to do and use AI to do more of the better work. And so I think we should all endeavor. Mhmm. I think it's a good framework to to use. Think about in your organization, what roles fit into what categories, and aspire. to shift all of your roles from the democratization category over into, to the extent it's a spectrum, into the professionalization category. Mhmm. Yeah. And and I think I lost you for a millisecond. I think I lost your voice for a millisecond then. Okay. You're muted. There you go. back? You're back. Yeah. You're back. Oh, those Internet gremlins. I know. Right? Technology. That's right. Next topic. What separates organizations that are redesigning work around AI from those simply layering AI onto existing ways of working, what do minimum managers need, and skills authority and incentives to support that transition? Catalina, you first. Yeah. It's just like accumulating mistakes. So when you layer AI onto old habits, it's like it's like putting a jet engine into a horse carriage. It's it's you don't get a jet. You just get a broken carriage. So, The technology itself is not going to do it. The true redesign means rethinking workflows around human judgment from scratch. I think the last twenty, twenty five years, we have been doing tech first. Like, what is the database? What is the engine? What is the compute? What is the machine learning technology that we're going to use here? All of that has been coming from this side. I think that today, that we already have that data layer, compute layer, intelligence layer. Each of those can be tuned to your own system to maintain it as reliable as possible, and that is what is rethinking the situation for AI today. You cannot build it into error, error, error. You have to start with a framework that is going to empower these SMEs not only to leverage the technology, but also to leverage the technology within the way that you work together. And and that's a very important caveat because it's not you in your silo trying to fix the world. It's it's you with your skills and your domain expertise contributing to the workflow that now incorporates perhaps somebody from IT risk, somebody from my attorney team, from somebody from my my new governance team. So twenty years ago, we were not working together towards this workflow. I am a data scientist, and I can say that a lot of us just sit down and solve the problem and go deep into the weeds and take and solve and and then so what? Then now what? How do I deploy this? Who is gonna use it? How is this connected to my business use case? How I am proving that this is generating value. All of those extra variables is that we need to put together here. Layering AI gets expensive. It's going to give you a small gains, but empowering managers to redesign work builds an agile organization where the tech does the heavy lifting, but people drive the purpose. Catalina, you could drop the mic after that comment. That was wonderful. Lest we get caned off the stage, we have a couple minutes. We're gonna try to get to a stopping point here. We've got one more important topic, and then I know we wanna wrap up and get everyone out of here on time. Most organizations measure AI success in productivity terms, just time saved and tasks automated. What should leaders be measuring instead? Yeah. We do have a lot of metrics that correlate with success that are part of that operational efficiency category. Yes. Something that it took two weeks now happens in twenty minutes. Those things are very tangible, but I think we have to go way beyond the time saved here. Think about efficiency metrics also, where the capability is being measured. So for example, value velocity measure how fast an insight, for example, becomes an actual business solution or a business decision. Not just how quickly a draft is generated, but the entire cycle. This is in production. This is maintained. The capacity redeployment, for example. Track where you are saving time, but it's also how is this being deployed? What is the strategy, the customer impact, the mentorship? Are designing a new workflow? Is using technology? It's using innovation. How many people knows how to do it within your company? It cannot be silo innovation, has to be from the top down because we have a new way to do things, we're going to do a capacitation. So we are gonna go get trained and understand how to do it. Right? So this value velocity, capacity, redeployment, and then your judgment, quality, and trust. If you cannot sign your name on that AI outcome, you don't have the trust. And to have that trust, you need to have transparency. And to have transparency, you need to have traceability. So it's not only the tech for tech. Is the tech for tech with my human judgment, with my framework and my guardrails, but at the same time, I understand that my tests, guardrails, and everything that I need to sign this off is in place. Therefore, I'm gonna support this outcome. Right? So efficiency is baseline. Empowerment is the ROI. Beautifully said. It's time to rethink performance management. I firmly believe even though these feel like soft metrics, we need to redesign our incentive systems. We need to redesign our leadership mechanisms. I think that the. the way we should measure the value of intelligence as a service is time to insights, time to inspiration. We should think about the quality of engagement. And, again, although those seem like soft metrics, the onus is on us to figure out how to culturally incorporate that as an ethos into our organizational DNA. One tool that we we use frequently at People Rain with our with our customers, we call it taking the pledge, and I find that this, Mhmm. is a way to encapsulate a lot of the important topics Catalina and I have been discussing. The pledge is very simple. It goes like this. We ask every employee to, make this commitment. They say, I will openly and enthusiastically share how and when my work product was influenced or augmented by AI. It is a profound a profound impact on a company's. culture. Mhmm. It takes AI out of the shadows like we talked about earlier, and it gives you kind of a framework where you can share with everyone in the organization at all at all in all facets of the of of the org chart that, we're doing this together. And so when I. talk about some of those new metrics that we should be using to measure the value of intelligence as a service, it starts with something like the pledge that shares that culturally. We're investing in a culture of innovation. Catalina, I I know we're out of time, and there were a couple other things that we wanted to share with the audience maybe. I don't know if, you or if Renata would like to share about this event that's coming up where we'll we'll be having more of these conversations. Yeah. I I I would love to invite this audience. I mean, for me, speaking of the importance of the human right now in this era of AI, for me, the three dimensionality is very important. And this event is gonna happen in person in New York, and you can just, take a picture and scan it. This is gonna be in September 24. And the why I think is so important is us talking to each other, sharing what's working, what's not, how you fix something that was so hard. We are all leading this moment and this is happening. It's a chapter in humanity and I think we have a beautiful opportunity to really embrace it and change things and innovate and do better. But at the same time, we have to be very aware of the behind the scenes and how bad outcomes can actually bite us back. The invitation here is, let's enjoy some time together. This is a full day where customers are sharing how they are deploying AirViews, how the framework is embedding that governance, what's working, what's not, and opportunity to do hands on exercises within Dataiku as well. It's a very nice opportunity to be together as humans defining this new chapter in humanity. Definitely invite everybody. We have a few minutes. I think this is where we can actually open for Q and A. I have been checking some of the comments as well. And I think we have a very spot on audience that has been going through this evolution. Like, we know this is happening. I will say that every major technology in evolution has changed the task before it changed the value of the human. So you mentioned steam then. So steam amplified our physical power. Then in the nineties, we have Internet. Internet amplified our access to information. Now we have AI. AI is amplifying our cognitive power. And every time the organizations that won, were not the ones that protected people from the technology, the opposite. They were the ones that empower people to do something bigger with it, and that's where we are today. Catalina, Idris asks, how do you align executive demands for thinking about changing? See. I believe there's more. Changing the overall structure of my data analytics team. I guess. maybe that's kind of a day a data accrue question if you wanna answer that. Yeah. It's it's an education. It has to be a little bit of education involved because not everybody is equally aware of the potential bad outcomes that AI can generate within your workflows. So I lead a lot of workshops where we bring the leadership teams together, and for the first time, we think outside our own bubble. The invitation here is, think outside your bubble. You may be leading this initiative, A, B, and C, but innovation and the proper use of AI so your humans are as empowered as they can be, has to come from the top down. You have to have a vision. By the end of the year, I'm going to have 50% of my workers certified on A, B, or C. You're going to measure what success means to you. By the end of the year, we're gonna change these three or four workflows that were completely messy and completely drafts into production ready with guardrails and governance embedded. By the end of the year, we're gonna have, legal and governance team beautifully included as part of that framework. So it's not only machine learning, it's also an outcome that needs to be responsible, transparent, trackable, all of that. No? So it's an invitation to think bigger and to think outside your bubble and to ensure that you follow that mission, ambition that comes from the top down and you are part of it. This cannot be inside us because then you're gonna go nowhere. We're gonna take one last question and wrap up. If we don't get to your question, first off, sorry. We had a lot to cover, and this is certainly more than a, a forty five minute conversation, but we'll pick back up maybe at the live event or in subsequent sessions. Catalina Sophie says, what about the application of leadership to AI? This is part of the process. No? You need to and sometimes it's the chicken and the egg. I have seen that too. Sometimes you create something that it was a pilot and it has humongous potential and you go from there. What do we need to do to embrace something like this? But it goes back to the education. Understanding what are the risks or the extending, what are the potential outcomes and being very aware of that. And that's different from ten ten years ago, fifteen years ago. For us, tech tech is tech. Tech is tech. Right now we have a generative AI layer, and that brings non deterministic outputs. And what that means is that your human has to be more involved than ever to ensure accuracy and to ensure that you have that trust because without trust, you have nothing. It's an exciting time, and the fact that you all chose you voted with your calendar. You chose to be here and be part of this conversation and part of this community. You've taken at least the first step towards saying that I embrace the idea that humans plus machines, you to the power of AI, is the future that we all wanna be investing in. Catalina, wonderful work. Thank you to you and to the team at Dataiku and for the great research that you put forth. Let's make sure that this is the start of a dialogue, and, thank you again to everyone that joined us. Thank you so much. It was a pleasure then. See you around.