Name: Beyond Aspirations: Building the AI-Fluent Workplace
Uploaded: 2025-11-05T19:03:05.152Z
Duration: 47 min 12 s
Description: Beyond Aspirations: Building the AI-Fluent Workplace

Transcript for "Beyond Aspirations: Building the AI-Fluent Workplace": Okay. Hello. Good morning, good afternoon, and good evening. Thank you everyone for joining us today. It's great to see so many people here from around the world. My name is Giles Douglas, and I'm the CISO here at Grammarly, which as many of you may have heard is now part of Superhuman. Last week, Grammarly announced some big news. First, we changed our company name to Superhuman. The Grammarly product you all know and love will remain, but this new name represents our shift from a single product to a future suite of tools that companies and teams can use to truly capture the value of AI across every team and every workflow in the organization. We also believe that AI should not replace people, but help them do what they do best. So our new name reflects this mission to build AI tools that extend the capabilities of our already very talented employees. You'll all hear more about this evolution from us in the next weeks and months. But today, we're not talking about our product. We're talking about how people, leaders, and organizations can develop the skills, environments, and cultures that our businesses need, not to just use AI, but to use it well at every level in the organization. We've had over 5,000 people register for this webinar today, so it's obviously something that we all have been thinking about. And I look forward to discussing it with my esteemed colleagues here, who I'll introduce in just a second. So before we start, a few bits of housekeeping. Because of the size of the webinar, all of us, other than the speakers, are automatically muted. You can enable closed captioning. There should be a CC button at the bottom of your webinar window, which is pictured here. We would also love to hear from you throughout the session. So please leave comments in the q and a widget. We will have the team working from behind the scenes to help answer your questions. And finally, don't worry about taking notes. We're gonna follow-up with a recording of the event so you can rewind and rewatch to your pleasure. You will not miss a single thing. So ah, wrong slide. Alright. We're gonna cover a lot of ground today. So after the last over the last couple of months, we've been rolling out an educational series on AI fluency, which we'll link to in the chat now. My co speakers and I joined that for a number of fireside chats and shared key steps organizations can take to scale AI fluency in their organizations. And today, we're bringing the band back together to dive deeper into details. So over the next forty minutes or so, we're gonna dive into a number of topics. What is AI fluency? Why should you care about it? What are the building blocks of AI fluent workspaces? How we can drive personal AI fluency? How do we, as leaders, create the right environment that facilitate fluency? And how do we create cultures that inspire the workplace to innovate responsibly? And if we have time, we'll also take questions from the audience. Now I'd love to introduce the rest of our speakers today. So firstly, back to myself. I'm Giles Douglas. I'm the CISO and head of infrastructure here at Grammarly, also known as Superhuman, and I spend most of my time managing our security infrastructure and lately deploying our organization wide AI strategy. I have a unique perspective in doing this. I'm both a builder and buyer of AI, so I'm gonna talk about both today. And it's my pleasure to be joined by Libby Rodney and Tim Sanders. Libby, would you mind introducing yourself to our audience? Yes. I'm Libby Rodney. I'm the chief strategy officer of the Harris Poll, and my job is to sit over research and thread the needle between macro insights that are happening across across culture and business. Great. Tim, could you introduce yourself as well? Sure. I'm Tim Sanders. I'm chief innovation officer at g two. I lead transformation of the business as we move from the SaaS era of software to the agentic era of superhuman strengths. Great. So what are we all doing here? So I'm sure you've all heard of AI fluency before. That's why everyone's joining us today. But I wanna start by giving a simple definition of what we mean by this. So we mean knowing how to use AI effectively, which is obviously more complicated than it sounds. It requires you to understand how AI works. So you can not only just evaluate its outputs, but you can use those outputs more effectively, and you can use them responsibly. It goes beyond knowing how to just prompt a chatbot, but also knowing how to integrate AI into your workflows so you can apply it to solve your real world problems. As we all know, there's many different skill sets, there's tasks, there's decision making processes that happen all the time across our workspaces. So it's critical to help our entire workforce be confident stewards of AI to get real value out of it. So why is it really now important now for companies? Tim, you're literally writing a book. It's called AI fluency. Can you talk to us a little bit about why it's so important for people today? So when you think about AI fluency, it's your understanding of not only the systems of AI at just a use level, it's about the mindset of artificial intelligence. When you realize that AI is just a prediction machine, it takes information that you have, produces novel information that you don't have, it's a breakthrough in personal productivity. But it's important because it will impact how much money you make and how far you move up the career ladder over the next few years. Recent research by Indeed indicates that if you can demonstrate in 2025 generative AI competency, your ability to leverage those tools to produce more results than the average person, you'll see 47% more income in your job placement. Also, if you take the time to learn AI systems and mindset on your own, you are three times more likely to be promoted than those that wait for their companies to hand feed it to them. Because at your companies, AI will only become more strategic. My favorite analogy I've heard recently from BetterUp is that you should think of the concept of, are you a pilot or are you a passenger? Because, folks, this plane is taking off. In today's AI, I call them background agents, they make suggestions. Those suggestions are valuable to your productivity. Very, very soon, the tipping point will be obvious where we'll have the action agents, and then you'll need to be a leader of agents. So think about it this way. Today's AI fluency is tomorrow's AI trust. And if you learn how to trust action agents two, three years from now, you'll literally be 10 x more powerful than the person sitting next to you that doesn't. Correct. So, Libby, you partnered with Grammarly and the Harris Polls to develop our twenty twenty five productivity shift report, which has us uncovering a bunch of key findings about AI in the workplace. Can you tell us how AI fluency is transforming productivity? Sure. I mean, first of all, this these are some of the numbers you see in the middle of the slide here. So 37%, the more time saved by AI fluent workers than less fluent partners. They're also seeing, and it improves their relationship with their customers at 9%, and it improves the quality of work at 12%. And 15% say it's more likely to increase work satisfaction. So we're seeing really strong numbers when you have fluency. The challenge is or, you know, opportunity and challenge is only 40% of the workforce identifies as AI fluid or AI literate. Now that means that AI fluency is really only happening in pockets of the organization, and yet you're having these high gains. We also see in our work with OpenAI that the not only is the if you have literacy in your in your, business life, is it great and you're getting these gains, but in your personal life, it's great. We've seen that 65% of millennials say that, using AI tools opened up worlds that they never expected to be open and made things achievable that they never expected to be achievable. So people are experiencing gains both in the workforce but also at home life by understanding and really adapting to AI, you know, similar to what Tim was talking about. You know, I have to just plus with you on one thing, Libby. I, I I heard a professor at HBS, Kareem Lakani, talking about the idea that when you talk about a renaissance person, so you say, oh, Libby's a renaissance person. What does that mean? She can paint. She's a poet. She can present. She can play guitar. Those are all skills that people have to invest a lot of time and effort to learn over years, So that's why there are not a lot of renaissance people. But to your point, with AI fluency, everyone can be a renaissance person because it shortens that that that learning curve for acquiring new and novel skills. Yes. That's right. And then in other research we see though, and I think we'll talk about it in in the context of this conversation, is there's so much fear and anxiety about which skills to learn, how to learn them, to your point, Tim, like, you need to do it for yourself versus waiting for the company to do it. And there will always be early adapters and super adapters who will win in these, evolutions, but it's also about how do we get everyone to take the journey with us because Yeah. There are some people who are inherently don't know how to take that journey, don't know what skill to start with. And I think that is also an exciting time to be a leader Yeah. And champion so many people to get off the sidelines and get into the AI game. That's it. I think activation is the key. I always tell people, like, what's your Achilles heel when it comes to deadlines? Like, what's the thing you're always running late on? That's your starting point, folks. That's the easy way to get started. That's how enterprises are getting started with agents. They they start with the pain. Right. And I think we're seeing that in studies. Right? So we have all of these studies that have taken place that show people are investing in AI, but they're not getting good value out of it. And we also see this kind of bizarre situation where if you ask the c suite, they're all like, my employees aren't ready for this. But as you as we're pointing out in this conversation, people are ready. They just don't know how to use AI to solve business problems. So it's not integrated into your workbooks. If you're cutting and pasting to a chat tool, it's not gonna really improve your productivity that much even though, obviously, you're using tools. So in a lot of cases, we're measuring the wrong thing. Yeah. I think so. And and, you know, the research is real on enterprises not seeing the value pull through. Here's here's an idea. There's a concept called Parkinson's Law. If you haven't heard of it, go ask chat g p t about it. It'll blow your mind. But, basically, the concept proven over seventy years. Human beings always expand to the time that they're given to finish the work. So what happens is if you experience productivity, you also experience the luxury of time. If you're a perfectionist, you obsess over perfect, and you still take the same time you would have taken. If you're not a perfectionist, you watch TikTok videos. You get distracted. You do other things. The trick now is how much do we advance fluency so we give the agents autonomy? Because until we give the agents autonomy, Parkinson's Law is creating this last mile problem between individual productivity gains and actual business results. Yeah. So I'm gonna move us on here. I think there are three building blocks to fluency that we're gonna talk about today. We start with personal fluency. We need to have our employees be confident using AI. They've got to experiment with it. They've got to actually translate it into measurable impact. We need an environment from a business that facilitates this. We need to give people the right tools, the systems, and the guardrails to make it safe, practical, and accessible. And we need a culture that inspires fluency. We need people to understand that all of our teams, all of our leaders are using AI, and we want everybody in the companies to benefit from this and really bring out the best value of themselves. So our goal today is to explore each of these areas and how your teams can move from isolated wins to enterprise level impacts. We know today that we have fragmented adoption. We have productivity hotspots. We've all met the person who is absolutely great, and you're like, woah. How could I actually be that good at using AI? And we see others that are struggling with that. So we're gonna talk about how we can actually make this better. So this all starts with personal AI fluency. The aggregate of each individual's employees impact is what drives organization wide outcomes. Tim, can you walk us through the path that you see to personal AI fluency? What can people do concretely? So, I wanna use two analogies. First of all, think about physical fitness. People ask me, what do you need to do to get fit? I say, get up and move around. So the secret to AI fluency is to make a time commitment every week to how many hours you spend in tool. That's what I do. I spend I spend myself personally ten to twelve hours a week in the tools. You don't start there. You start simple. A couple of hours a week. Track it. Think about the book, Atomic Habits. So make this a part of what you do. Here's a cheat code. Use new AI browsers like Atlas or Commerce. That's a great way for it to be sitting right there, plain and obvious. But that's the first idea. Start using it. But you also need to do exactly what you're doing today. Step back and study it. Spend time researching it. Read books like Co Intelligence by Ethan Malek. Listen to podcasts like the AI Daily Brief, by the folks at Superintelligent. All of those are gonna give you more perspective on how the tools evolve. And the final thing I'd say, Giles, is the most productive thing you can do at work is share how you won with AI with other people, not your use cases, not your cheat codes, how you achieved superhuman results and work backwards. The research indicates that as you share how you used a technology like AI, it actually sharpens your skill exponentially because of the feedback loop and your ability to enunciate your intention and the outcomes. So think about those three things in tandem. And here's the last thing. Think about stacking skills. We talked earlier. Where do you start? Are you gonna start by using AI assistance to improve how you write? That's how I started. Are you gonna use it to keep up with the flood of emails in your inbox? I know a lot of people that do that. Maybe you use it to make charts. Maybe you use it to do some vibe coding, do prototypes. Try to add a new use case at least every other month that you seem moderately good at because learning effects, just like learning foreign languages, learning effects says every time you do the next use case, it's easier and it's significantly more valuable to your career pathing. So take advantage of learning effects today. Hit the AI gym and make sure and spend meetings talking about how you won. And I emphasize that because if you don't talk about the business result you got from AI, other people that aren't on the bus yet might think you're lazy, and you don't wanna do that. Yeah. I love that. And I think that really speaks to me as an engineer. We always find an engineering iterating works really better. You can't just expect to be an expert from day zero. Libby, what have you seen? The conversation about AI and skills is an interesting one. In some of our data, we've seen that, you know, globally, 76% of workers say that AI will create new skills that don't exist today and that are needed for employees to be competitive. We are we see that six in 10 employees have already created new skills, that they've learned from AI. So there's I think what's important to recognize is that we are and will continue to be in this era of disruption around AI and what skills that you have to apply. And so the when we talk to people who are thinking about this from their personal point of view and the future of education, AI, and skill set, the first thing that you need to learn and and adapt, and this is exactly aligned with what Tim is saying, but is learn how you learn. So approach AI how you best. If you are video, audio, visual, do that and then, share that. So when you learn like, when you do it and then you share it to Tim's point, like, that just, like, actually creates expertise. And what's so exciting about this moment is because people really don't know what they don't know in so many cases, there's such an opportunity to learn a small thing over here or over here and become an expert in it and become, like, celebrated for what you can bring to the table. So that's a really once in a lifetime exciting moment. I think we're we're there, and it all starts with how you learn best. So I think that's great. That's great. Chase the yield. That's the way to think about it. If you get the greatest yield by watching video, that should be it. I I know a person I work with at g two. She's getting the best yield by making it part of stand ups that little ten minute show and tell is teaching her more than listening to podcasts and reading books. That's where she's getting your her yield. I think that's a good way to think about that, Libby. Yeah. I sort of studied the other week where in an educational setting, they found that customizing the course materials the way the student learns really yielded huge results in them actually understanding. I think it's kind of a thing we couldn't do before. Right? It was impossible for us to go make 50 different courses to learn the same thing. Having a tool that adapts to you really makes it work. It's not just novel. We're learning sticking power. We have good AIU tool usage. So let's keep going. Getting comfortable with AI is the first step. Right? But fluency isn't all of it. It's not just knowing what AI is. It's about how to apply it in meaningful ways. Mhmm. So right now we see that too many organizations kinda go, I can bolt on AI to the end of this. You throw a chatbot into a web page. That's really limiting the potential of doing this. So we need to reimagine from first principles, not just how can I tack AI on, but how do I rethink the process? So, Tim, you've done some work thinking about use case discovery. Can you talk to this to us about this? Absolutely. So to your point, think of a skill as your ability to do something well. So when you're building AI skills, you're improving your ability through AI to do something really well. We talk about atomic blocks. I I mentioned this book before. Atomic Habits, great book. And when you think about the idea that you wanna make it obvious, you wanna make it rewarding, you wanna make it easy, it's really about integrating the artificial intelligence tools into your existing workflow, so guess what? You can do something very well. So this example you see here, would be, what a lot of people have as a remit, and that is to be thought leaders. And for in this case, the objective might be drive reach on LinkedIn. And you say, well, what's the deliverable? Well, I've got to produce content, whether it's a video, whether it's written content, etcetera. Then what you need to do, and this is where it becomes atomic, is you taskify the deliverable. So I begin to simplify the concept that there's seven tasks in delivering thought leadership content on LinkedIn. And then the idea then is you go through and you challenge yourself to apply AI to the most places it can apply to. You can always decide not to later based on feedback. But you see here, I use it for research at the agentic level, meaning deep research. I even have an agent called Snopes that I built that checks deep research, and it makes my research loop very dependable. Check mark. Outlining. It's a conversational tool for me. I'm gonna work more than assistant on that because I like my own nuance, and sometimes it's a little bit obvious or general. Remember, AI is a prediction machine, and it predicts the median, not the exceptional. So you have to think about where you wanna converse. Writing, thumbs down for me. I use AI voice dictation, and then I use a tool like Grammarly to improve it in a copy editing mode. Recent research, that I just saw two weeks ago. You are 35% faster using AI voice dictation and a writing editor than you are using chat g p t to author the content if you're accounting for natural voice and quality of work. You can optimize it, meaning you can make it better. I always use Grammarly conversationally. In other words, make suggestions. I don't have it actually fix things directly. Then you wanna be able to illustrate it. This is something I'm now agentic on. I'm a 1.5 shot person with charts and graphs, but it takes time. I am conversationally letting it post. I wanna let it be agentic, but there's high blast radius there. You can appreciate that, Giles. I I can't quite give it my credentials yet, but, hopefully, I'll I'll put a check mark there, but when we talk next year. And then I have absolutely used agents now to analyze, give me feedback on content cadence when it should have been posted, how it could have done better, etcetera, saving my managers some time that way. That's an example, and you can even see the sequence of how I built this. Sure. Let me can you give us your thoughts? Well, I'm just so impressed by what Tim is doing. And I maybe I represent, you know, someone else listening to this, episode and thinking about, gosh, I wish I had, like, the time and the week to to do this. And and, actually, Tim, what you're saying is really inspiring to me because it is it is like creating the ecosystem and the map and learning and tackling month by month, week by week. Like, where do I find value in this? And I think, ultimately, you're winning if you're just starting and you're getting curious about where AI can deliver some value for you and which use cases it can. That's right. Right? And then, Tim, do you think I mean, I'm hopeful. I so I know you do all this work yourself. But I'm hopeful that this work will come to us a little bit. Like, you know, we're at the beginning of this, but then products and services and solutions will just come to us more seamlessly so it doesn't have to be such a heavy lift. Right? I agree. I agree. I think that agents will be in a place in two to three years where where the superhuman skill is to express goals clearly, provide context at a granular level and be a very quick evaluator. And so you'll be in the future, you know, you won't really worry about how to use the tools. You worry about the clarity of the goals you express to agents, the context you give them to operate with high efficiency. And as Andre Karpathy would say, how fast that evaluation wheel spends. Because again, think about Parkinson's law. If it takes you too long to approve actions, you're the bottleneck. Yeah. I really I Yep. And and a lot of this comes down to we're making everything into a specification problem. You have to teach yourself Yep. How to specify a problem succinctly and repeatedly. And that's a skill Yeah. That a lot of us don't have yet because we're not used to using these tools. I would say one thing that Libby brings up a very good point. Like, where do we find the time for this? Like like, really. Twenty five years ago, I published my first book, Love is to Kill Our App. It was about the idea that you should read two books every month. It was the biggest question I always got from people. Where do I find the time? The answer is, covet it so much that you start saying no to other things that don't compete with it from an ROI basis. In your life today, how much time do you waste while on LinkedIn? How much time do you waste going to Zoom calls where you could politely decline and watch back the meeting later at the gym? Think about all the things on your calendar that are robbing you from your ability to learn AI, covet the AI time, block it out in your calendar, just say no. I am really excited about this future of, like, just talking, whispering into AI, and gaining those efficiencies and then being able to really spend the hard time on the critical thinking. You know? I mean, I think for those of us who love critical thinking and ideas, like, this, again, couldn't be a more exciting time in in that, light. Yeah. And if you look at this, the key is not just using AI. It's using it in workflows. There's a lot of studies out there. There's another one from MIT that shows 95 of AI pilots fail, which conversely means that five percent succeed. And, ultimately, we think they have one thing in common. They actually do something that's useful. They integrate it into your workflow. And that's not always the same for every person. It's not the same for every employee. You have to understand what workflows matter to the individual and integrate AI in ways that works for them. Well, I would say I I have to just jump in really quick because anybody that follows me on socials know how crazy I am about this MIT study. Really important. It's done by the Nanda Institute. It's only about generative AI custom projects. The word agent doesn't appear in the entire report. And what they were what they were talking about is that you can't build a custom large language model and actually see P and L impact in that two year window, that you've identified for ROI. Now what that means is you can't chase last year's innovation. All you can do is kind of move forward. So that to me, that's really the pull through. However, to your point, when I'm seeing organizations at g two that report high satisfaction at a ROI level, the one thing they all have in common is they're not talking about agents, they're talking about agentic workflow. So it's not about having one off tools that are point solutions, it's about having a systems mindset. I just read a great book. Here's a hot tip for everybody watching. It's called Reshuffle. Reshuffle says stop thinking about AI as a replacement for people or for that matter, an augmentation for people. Think of it as a new system of work, like the metal container and how it changed logistics years ago. It changes at a system level how we should be thinking about doing work. I believe in the future, what we'll think of AI as is a coordination tool that takes everything around us and makes us superhuman. That's I think what's coming next, but that's what workflow integration means. How are you orchestrating everything you do to do it well? Yeah. I I think that really speaks to what we're gonna talk about next, which is it's not just the tools that use it. I mean, everybody, if you look at, your customer service app, your sales app, everyone's adding a chatbot into these things. That's not really getting to how to make AI useful. We need to make sure the tools are available at the right time, and that kind of implies that we have a platform that draws all of these things together and provides you with a mechanism to work with them. And to do this from the security side, we also need guardrails. But our guardrails can't just be don't do that. We have to provide, like, a paved road for people to be able to walk along it and follow something that actually works for them. The the guidelines that we have to do need to provide safety, not just limitations. What have you two seen from this perspective? Have you found that there are ways of utilizing AI which is enabling not don't do this? Go ahead, Libby. I was just gonna say, I mean, we do a ton of research in the space with CIOs, CTOs, cyber clients, but also then on the other side, the business leaders and CMOs. And this is the big tension, right, of, like, don't do it and people work around and do it anyways. So I think the the challenges that we're seeing at at a corporate level is that we're in this very gray zone where you do have to prioritize security, but, employees themselves or business leaders themselves are making choices outside of that, and that actually puts you at a higher risk. So that risk is is on the table without an actual understanding of how do we we are building the plane, but, like, how do we do it line by line so everyone's on board? Yeah. So to me, it goes back to the the idea that this is a journey that the organization is on and all the stakeholders have to be part of it. That's a great that's a great point of view, and I have two two things to add to that. So to your point, Libby, people will use their own things, but they use their own things because the thing you gave them was neutered and it doesn't work. Point 5%, that's half of 1% of people in our recent survey use their corporate provided chat GPT the most. 47% use chat GPT, of which 70% use personal. Do they wanna be shadow? No. The one they were given by work has too many guardrails on it. And as a result, it's not performing, and it's not going to compete. So you are always competing with general purpose tools in the real world, so keep that in mind. So that leads me to my second point, safety without friction. That's how I think about it. Now, Charles, you and I probably have some have a beer and really talk about that a little bit more. Here's what I mean. I think, and my some works that hack the box, so maybe I'm just a little bit, biased about this. Culture and training is really important to this. Right? So one of the most important exercises I'm going through now with people in my organization is this concept of the Jeff Bezos one way door, two way door. Mhmm. So I think a lot of the things you approve in AI are are are two way doors, but we labeled them as one way doors. Meaning, if that if we do that thing, we can never take it back. Look. Right? When you think about IT operations using agents, for IT remediation, they let it rip and they do instant rollbacks and daily QA because those are two way doors. The debt is worse than the impact. So what I'm saying here is, as a leader, be very realistic about the AI risks that truly are one way doors, a bad reflection on your brand, losing a customer because of sending the wrong invoice, but not everything is. And you should have a goal to follow that reliability curve so that if you start out today with ten one way doors and a workflow, by this time next year, there's only two. That's how trusted option is gonna work in the future. But I think it's about coaching, and I think it's about us stepping back and saying, how do I get safety without friction? Yeah. I think you need good off ramps for risk areas. So close your eyes. Customer personal information into your personal chat GPT. So provide a mechanism such that someone who wants to, phrase an email or work with a customer effectively using AI, but does it in a sanction tool. So don't just say no. Say this is how we want you to solve this problem. That's right. I agree. And and I give you an example. I I mentioned I'm just so I'm just so bullish on, AI voice dictation now because it's fixed. And you look at these tools out there like Willow Voice or Type list or whatever. Our info security gave me a really good point of view though when you think about this, Giles. It needs to be local. Why? Because we say things and we don't think as studiously as what we upload. Well, I didn't think about that. So that's an example of how the tool integration works like you need to think about what stays on your computer and what goes to the vendor's cloud. I think that's a good example too of of really figuring out not just what you upload, but what leaves and goes on the outside. And it's pretty insidious how that can work. I I think providing examples to your teams is really useful here. And those can be positive and negative. It's totally fine to say, yeah. I tried to solve this problem with Gen AI, and it didn't work. And I think I think admitting some failure in using the tools really helps people explore how to use them and encourages Yeah. Discussion. Yeah. I I love to say I tried it and it didn't work. I love to say that all the time about hiring ex Google people. Hey. I hired an ex Google person. Didn't work. Like, really? I can't believe that. Lily, why don't you think a lot of them now. Well, I I would say that, one thing that I find is really interesting, especially when we're talking to CMOs, they are getting tools for their teams to use AI, and their teams are very frustrated. Their teams are not talking about them. Mhmm. So I think there's also this this silence as a deadly killer from an organizational point of view where if you're just talking about it and it if only teams have even AI in silos too, and there's not this, you know, top down or top down transparency and bottom up experimentation mode that's happening in the organization, you still kill it. Like, you can't act alone in this kind of island within your organization, and that that has to foster more openness around what is really happening. I agree. I agree. I I I think you also need to incentivize it in a positive way. So two things that I'm really on right now. Stop talking to people about cost and time savings. Time savings means headcount cuts. The easiest way to kill AI adoption is to say how much money did we save with AI, how much man hours did we save with AI. No. Talk about winning with AI, beating the competition, three x in our cadence of marketing outcomes. Focus the language if you wanna get the adoption because I'm seeing that as a, you know, okay. The CFO, the chief frugality officer can have that point of view. Let's keep that behind the firewall, but but watch watch your words. The second thing I would say is speaking of words, we've gotta make it part of the, career pathing process. What do I mean by this? Mhmm. I will no longer hire anyone who does not have AI fluency at the point of the interview, and I tell all my people that. No. We did not go with that candidate. Why? Because they're just getting started with AI, and that means they're not committed to it. I don't care that they went to Harvard. I don't care that they worked at this great company. That is a ripple effect in your organization. Everybody hears that and says, wow. AI fluency is gonna be part part of my focus next year. Now that's not punitive. That just helps them understand this is now becoming, at least for me, 30% of a person's skill potential. That is important too to couple with with focusing on augmentation. Yeah. And I I think it's good to the last point on the slide here. We need feedback loops, and things change. I I have a lot of, time seen engineers doing interviews where they'll be like, I think this person was cheating because they used AI in the interview. And I think it's really the wrong framing of the problem. It'd be, well, how are they using AI? And wouldn't you expect them to do that in their job? It's like an interview is an artificial situation anyway. We want people to be best possible give them the best possible opportunity to show what they're good at. And if they're using AI, tell us how they're using it. And if we can get people comfortable with that, then we'll have to absolutely go towards what you were just saying, Tim, of Understanding how people use AI to solve problems and encourage them to show their best sides here. That's that's a I love that idea. Bothers me so much. Oh, gosh. I'm sorry. I just gotta jump. Like, from academics to the workforce, it it to your point, Gail, it it's about how do you use it, how do you critically think, and how then how can we understand your mindset from that? I've had interviews where I've interviewed candidates, and I'm like, oh, did you use AI? They said, yes. And I said, what did you do? And they went through, like, two steps. And I was like, that's why this is bad. It's not the fact that you used AI. It's the fact that you only took two steps to think about it, and I'm getting better results through a one better prop than you. You know? And so I think there is just something that really comes to that that we have to get over the cheating element of this because it just holds us back in this, like, this mental jail, and it doesn't allow us to actually get through it and think about it. It's not the case where we want the one to run go you say, give me a report on something. I I'm not expecting five minutes later a thirty five minute, 35 page report that came back from an IG prompt. That's not useful. No. Sharing a prompt might have been useful, but we need to get people to read, understand, and make themselves better by using, AI in this way. I think we've gotta coach people. Right? We say there's a difference between a souffle and slop, and the difference is editing. So so use AI, but use it thoughtfully so as not to create work for the person you give the output to. I tell that to people all the time. Don't let AI produce work for other people. That's what they mean when they say AI slop. It's not edited. I think your problem is worse than the thinking. This moves into what we wanna finish up with today. Sort of, how do we use culture to inspire inspire fluency? If personal skills are the fuel, the environment's the vehicle, culture's the road you're driving on. Without the right culture, stuff's gonna stall out. So, Libby, you have your pulse on cultures that inspire innovation as well as those that don't. Talk us to us about what you're seeing today. I mean, I think, the first one and the first macro thing they recognize for all of us, and those who are hiring and running organizations is that the cultural and collective fear is high. And it's high across the board because of economic uncertainty, geopolitical uncertainty. So AI just adds more fuel to that fire. So the ways that you get people over fear, though, is that you add and you help them understand what the future could look like, what visions could look like. Tim was talking before about, like, what's the use case? It's not it's it's probably not time and just effort and productivity, but what are the other things that you can get better at, collaborative at, and the impacts that you can really make in an organization? That is going to get people over the fear. I think the the challenges, right, and we talked to, CHROs all the time about this, is that they're no longer even to to Tim, to your point about hiring on AI, they're not even hiring on AI skill sets right now. They're hiring on how adaptable are you and what your mindset's like because this is going to be everyone that we're talking to right now, everyone who's listening to this is going to be a lifelong student forever. And so being able to adapt to that, like, imagine the the vision of where the culture is going, but then also being able to adapt as a as a lifelong learner is going to be that thing that counters you from that fear metric that you might be feeling. So if you feel fear, your best thing to do is to start to learn about something. And that anxiety and fear will start to dissipate just a little bit at a time, and it will eventually go away a little bit more. So Solid. Solid. And and as a person who studies emotional intelligence, interest, which is curiosity, is a powerful emotion. It's actually as powerful as fear. It's almost like how gratitude eliminates jealousy in your mind. It's the same idea, Libby. So that as you develop a curiosity about the next thing, your fear about the last thing seems to subside. So that's just bang on for me, Charles. That bang on is kind of a nod for you, my friend. I always think about it too that, you know, I have to keep disrupting my my theory of excellence. Like, oh, I'm really a good prompt. Well, guess what? And I started this a month ago. I no longer write prompts. I only write meta prompts, and I'm really pressuring myself. I write meta prompts to get back markdown or JSON or whatever kind of actual prompt material, and it's dramatically improving my yield, my ability to get things in 1.5 shots. Well, that's been hard for me because meta prompting is its own bag of snakes. Right? Who knows what I'll challenge myself with later? Mhmm. Yeah. I think one thing that comes there is, the the tools are actually better at coming up with prompts than we are. So you have to get into this mindset of asking them what might be a good prompt to solve a problem. It goes back to what we prepared earlier. Specification is the hard part. Specification. The goal. That's the super skill of the future is to express a goal so clearly. Even a machine knows what you mean. And they I would have the critical thinking is incredibly important to that because if you're specifying the wrong goal, which is what we see a lot, or maybe it's my world of strategy, it's a Total. It doesn't matter. And I think that's that's gonna be so exciting for people who like to critically think. That's right. So I call it the Rick Rubin economy that we're heading towards. You don't need to know how to play all the instruments, but you definitely need to have a great sense of judgment and taste. Correct. So we're running out of time. I need to wrap it up now. So Okay. So lastly, before we we might have time for one question, but I wanna give a quick overview of how Grammarly and our new superhuman suite of products are helping organizations. AI fluency is a core focus, but it's important to get it right because the AI tools we have now sit outside of real workflows and require employees to change how they work. We've always taken a different approach. For the past sixteen years, Grammarly has worked right where people work and with writing suggestions that appear in whatever application or tool you're using. You don't need to switch. So in this evolution, we're going beyond just these suggestions and bringing an entire team of AI agents right to where people work. So this means that people no longer have to you remember to use AI tools. You don't have to remember the cute names, all those SaaS applications your IT department are deployed for you. Things will just appear in front of you. We will simplify the learning curve. So we're very excited about this more ubiquitous, proactive, and connected future, and we'd love to tell you more about it, how it can support your team or organization. If you'd like to talk to a member of our sales team, please click yes on the poll that you will see in your webinar window. So with that, I think we have time for one question to close this up. So let's move on to q and a. So in our conversations with leaders, what are we seeing as the biggest gaps between how knowing how AI works and knowing how to apply? Libby, what have you seen here? The biggest gaps in knowing AI and how to apply it are typically, just that actually, there is a big gap in terms of the age of the workforce. So younger people coming into the workforce actually know how to apply it pretty well. The leadership, obviously, you know, it it really ranges depending on their skill set. They tend to have really big ideas about AI, but they're not actually always going into the the barriers that are impacting kind of day to day work. So I think it's big vision versus, like, small experiments that aren't necessarily talking to each other, and that's a big gap that can be fixed by more organizational alignment. Great. Tim, can you close this out with a final thought? So it's not about utilization. It's about business outcome to your point, Libby. So we we keep saying we're gonna buy the tools and go find use cases for it. That is the wrong sequence of events. The gap comes because we're not starting with the business problem and working backwards. So we did a recent study on agents, and what we found out is the companies that did the best always applied agents to where the technical or the customer debt was the highest. Full stop. So it kinda gets back to something we talked about earlier. What's your Achilles heel? But at a company level, what's the thing you always run behind on? That's where we start. If you don't have the right vendor, go get the right vendor for that application. They're out there. Great. Well, I wanna thank everyone for joining us today. I hope you enjoyed this conversation as much as we did. As we said earlier, we'll send you a copy of these slides so you can go listen to us and have more thoughts. And, again, thank you for your attention today. Thank you.