🎧 Listen Now:

In today’s episode, we explore AI and modern work with Annie Dean, the VP of Team Anywhere at Atlassian, the largest distributed work company in the world.

Annie has advocated for flexible work for over ten years, stemming from her own needs as a new mom, but in this episode, we shift the conversation from WHERE we work to HOW we work. 

We discuss how to work more effectively, AI's profound impact on productivity and collaboration, its broader implications for workplace diversity and inclusion, and Annie's inspiring vision for the future of work. 

Key Insights from Annie Dean

Here are the actionable key takeaways from the conversation:

1. Default to Digital-First Norms

As Annie said, the way Atlassian works today has proven that the office no longer needs to be a dependency for how great work gets done.

Digital-first approaches optimize work efficiency, especially now that AI fuels even more seamless collaboration and information access.

Shift towards digital-first practices and utilize AI tools to enhance productivity and reduce dependency on costly physical office spaces.

2. Reduce Ineffective Collaboration‍

Fortune 500 companies collectively waste 25 billion hours a year on ineffective work, preventing people from doing the deep and creative work they crave.

F500 CEOs believe their teams can get the same amount done in half the time, but their people are wasting time being stuck in a system that they can't win against.

Annie’s study found that a lack of clarity on goals, a lack of knowledge sharing, and too much distraction are the three main reasons we are now getting our best work done.

When replying to a notification is more important in a company than progressing important work, or two teams finding out they were doing the same work after the fact, there’s something seriously wrong.

Appoint someone to solve this problem and put in the effort to fix it rather than worrying about where work will happen.

So step up your game in terms of how priorities are communicated and knowledge is shared, and help people create time for what matters to double your teams’ productivity.

3. Leverage AI To Further Boost Effective Work And Collaboration‍

AI can solve the challenge of information being in many people's heads or even digitally dispersed in various accounts, email threads, and Slack channels.

This especially works when there is a culture of writing. Then, it’s easy to ask AI what someone is thinking about a topic or where to find a particular document to continue working on.

AI can also provide strategic insights and simplify complex concepts. For example, Annie’s AI helped her explain Async in the simplest of terms.

Integrate AI assistants to facilitate understanding and communication of strategic initiatives and create a digital-first and writing culture to make it effective.

As Annie shared, teams that adopt AI are double as effective as those that don’t, so there’s still a huge opportunity for us all here. (See our recommendations for the best generative AI courses.)

Don’t be intimidated by AI, just ask ChatGPT something as if you were asking a team member, and you’ll see how powerful yet easy it is. 

4. Change Your Own Way of Working‍

Many of these lessons can start small. Working differently yourself can help you model the right behaviors to your teams.

For example, remind yourself that your calendar does not control you. Or trying out new hacks like dedicating certain days or dayparts to focused work, like Annie is experimenting with right now.

Then, sponsor your employees to do the same. Think actively with your team members about how to attack the work they need to do, for example, by telling someone to cut half of their meetings, as Annie did with a newly promoted team member.

5. Embrace Modern Work to Promote Inclusivity‍

As Annie said, “So much of the story about women's place in leadership and in their fulfillment and growth in their careers is about logistics. Their access to higher salaries was actually a question of logistics, not intent, talent, or drive.”

Make logistics work for everyone by making work digital and providing flexibility to accommodate the realities of people’s daily lives.

This way, we can fulfill Annie’s wish of having the people who have been left out of the workforce until now run our companies in ten years.

🔔 Available on:

Transcript:

Daan van Rossum: It looks so nice where you are. Can you share? You don't have to say the street address, but whereabouts are you?

Annie Dean: One of the best parts of being in a team-wide environment is the flexibility to spend some time outside of New York City during the summer. So we have a place here on Long Island. My dad is an architect, and my dad and I renovated this place together. It's such a haven of tranquility and calm, especially when we spend most of our days in New York City. So yes, I'm very happy to be here, and we'll be here for the summer, enjoying the beach.

Daan van Rossum: It looks amazing. I just moved back to the Netherlands. As you can see on my screen, it's dark and cold, and I'm wearing like 500 layers. So I'm looking at it. I'm thinking, should I have moved back? But yeah, probably. Okay. So we're going to be talking today about some things regarding work. And I want to start with two numbers.

The first one is 10. You've spent 10 years; I heard it in your running remote talk, and you mentioned it on Rob Sato's podcast as well. You have spent 10 years now talking about flexible work, remote work, hybrid work, distributed work, and teaming anywhere. Are you not getting sick of it? What keeps you going? 

Annie Dean: The reason I started this work is because I had my first son when I was 27. I was working on Wall Street as a big law attorney doing transactional real estate work. It was really fun in some ways and devastating in others.

One of them being that I would be at the office until 4 o'clock in the morning, and it was incredibly difficult to deal with being somebody who was relatively early in their career and so didn't necessarily have the autonomy to define when and where I was working from and needing to manage things like breastfeeding and seeing my son and all of those things. And it really depleted me in a pretty extreme way.

I'm a very strong person on the spectrum. So it led me to really consider: If this is how I feel, how does everyone else feel? And what does this mean from a more macro perspective? And when I started to investigate what things really meant from a macro perspective, I realized that, though much of the story about women's place in leadership and in their fulfillment and growth in their careers, their access to higher salaries was actually a question of logistics, not intent, talent, or drive.

That felt like a problem that just needed to be solved and was eminently solvable. That led me to really try to pick at the problem. And I've now picked it up from so many different lenses. Somebody who was a startup founder has done this as a consultant and has done this in multiple enterprise contexts now, and what's interesting to me is that it was obvious to me in 2018 that work would flourish regardless of where it happened from.

But now we have these evolutionary moments of the pandemic to show everyone that the office really isn't how we should organize our work. It's really not a dependency on how work gets done, but now AI is here. And so the way that we think about using technology and tools to facilitate work shows us that the better way to work is digitally.

Because that's how we can leverage all of this technology to its full capability. And when we work digitally, yes, of course, there are always going to be times where being in person is extremely valuable, and that should be a part of how we organize companies.

But we need to default to digital first norms, and AI is going to make that truly obvious to even holdouts who are still lost in the culture wars over where work should happen.

That inspires me to keep going because I started out solving a problem for myself. I do believe that this is a problem that requires a better way to organize society. I think that this is a watershed moment for inclusion in our workplaces, especially for women, for people of color, and for people who have disabilities, and it's just a way to get more capability and a better quality of life for people across the board without losing our ability to compete or deliver the highest quality work. So it's just so obvious that I can't help but make it happen.

Daan van Rossum: You're making it happen, for sure. It's really interesting to see the shift from, okay, this is really about where we work to how we work, which brings us to the second number, which is the 25 billion. We went from 10 to 25 billion. So share a little bit about the new research that just came out and the 25 billion hours that were somehow wasted at work. 

Annie Dean: For those of you who know the work that we do at Atlassian and the work that I've been doing throughout my career, I think research is a really important part of all of this work to help create clarity in this environment and put real numbers around what's happening out there.

So we have an incredible lab that we call the Team Anywhere Lab at Atlassian. It's run by my colleague, Molly Sands, who is a behavioral scientist. And we wanted to do research this year in our annual state of teams report that really captured the story of how today's workforce is feeling. And we had the instinct that people were craving a lot more time to do deeper work, whether that's solo or collaborative work.

They wanted to be more creative, and they were feeling completely overwhelmed, both by the amounts of notifications that were coming at them and by the reactive and performative nature of today's workplace, as well as just feeling like they were pulled in too many directions and forced to change routinely, like all we're talking about is change, and people are actually really fatigued about talking about what the future of something is. They just want to know how to win today.

So we went out to quantify some of those components, and we spoke to 5,000 knowledge workers and 100 Fortune 500 executives. And we found that people really are wasting a lot of time because they're stuck in a system that they can't win against. So what we learned was that 25 billion hours are being lost to ineffective collaboration in the Fortune 500 on an annual basis. And we got to that number because 93% of the Fortune 500 executives that we surveyed told us that their teams could get to the same outcome in half the time if they could figure out how to work differently.

And that's really profound because even when I talk to executives today, it feels like a really urgent problem to me to figure out how to work differently. And yet it doesn't even have an owner in most companies. There isn't a big effort to make these changes, and again, we're still lost in this. The prevailing conversation is more about where work should happen if that's an answer to these problems.

But I think this 25 billion-hour number gives us a pretty good justification to say, Hey, there are some changes that we should make to how work gets done.

Daan van Rossum: Molly and yourself, you've been doing research now for quite a while, study after study, finding the problems in the data and how we can solve them. 

What are some things in the latest study that stood out even to you, even though you've seen so much research on this topic already?

Annie Dean: Yeah, I think what we wanted to do was put numbers around what we intuitively knew might be challenges. We kept an open mind, but we wanted to know what was actually happening on teams. What are their day-to-day behaviors, and what do those day-to-day behaviors reveal about what's holding them back?

And what we found was that there was a lack of clarity around goals. The inability to access knowledge and a level of distraction that impedes progress are the things that are really holding teams back today.

So some of the most interesting figures are that this one kills me: 65% of people, knowledge workers, told us that it's more important in their company to respond to notifications than to progress important work. That's fascinating. And that speaks to the behaviors that are rewarding in today's environment.

Another thing is that about half of teams find out later that they're doing the same work as other teams. So their work is being duplicated by other teams across the organization. And then a very high volume of employees and people that we surveyed are telling us that it's hard to get the information they need.

I think there's really something to unpack around knowledge, because I think that's actually the core of the shift that we're making, which I could go into greater detail about. But I think those are the big things that are holding people back. Goals, making progress, and getting access to information and knowledge.

Daan van Rossum: Even just that 65% number, that's exactly what we're hearing in the return to office conversation, mostly in the US, I would say, but mostly talking about, if I don't see people, I don't know whether they're working, and therefore it starts to become important to jiggle your mouse frequently or respond to messages quickly because then you're “working.”

Whereas most of the time management research would tell us that no, actually, what you need to be doing is to be focused on a task, and usually to get good work done, you need to have a block of an hour and a half, or two hours. So speaking to Molly, I know that you've implemented some of those principles into how you actually work more effectively.

What are some of the recommendations that you would give to anyone, from a Fortune 500 company to a five-person startup, in terms of how to work better together?

Annie Dean: Yeah, I think that it's all about having the time to... First, it's about having clarity. So we want to make sure that we're really clear on the work that matters so that we can prioritize that work. Then it's about having the time to actually get that work done. So you have to have. Those two things are in concert. And then the third thing is that you have to make progress in modern ways that are really optimized for sharing knowledge with everyone and making sure that you're continuing to save time.

I can tell you about an experiment that I've been running on my own calendar. I was just talking to my assistant yesterday, and I was thanking her because she's so willing to go on this journey with me of trying different ways of managing my time, and right now I'm doing kind of Monday through Friday as full blocks, eight hours straight of meetings, one-to-one meetings, all of my leadership team meetings, anything that might be external that might be important, and I bookend the week with those realities.

Also in the afternoon, from about 3 o'clock onwards, it varies from day to day, but then I have another block of maybe 4 hours of meetings on Tuesday to Thursday. But I try to keep Tuesday to Thursday as a very open day. And that means that what happened was that I ended up actually being in Zoom forums with people for most of Tuesday through Thursday. But I'm able to tackle the really big issues.

For instance, we were trying to build a strategy to break through with a new set of partners. And I spent an entire Tuesday looking at our organizational charts, thinking about our strategies, figuring out how to make that plan, and designing shared goals.

We are working on a big overhaul of our onboarding right now. So I'm able to go extremely deep on that activity and not just give my team a half-hour of guidance. But really dig into every assumption and uncover where there's opportunity. And by giving an activity 6 hours of my time, that is a very different level of investment than a surface-level kind of skipping the stone across the water form of leadership.

It's much more collaborative. It's much more rewarding for me. By the end of the week, I'm so fulfilled because I can look at my five big priorities for this entire fiscal year and say, wow, like I'm really investing where I think it's important. And that, to me, is a great way to organize.

And it's a matter of continuing to use that formula and changing it. Do I know what matters? Am I actually spending my time on it? And am I progressing with what I'm doing in a way that creates the maximum benefit for visibility for others, saving time, etc.?

Daan van Rossum: Yeah. So it sounds like you have some principles, but you're also very willing to keep experimenting and keep finding better ways to execute on those principles, including these big blocks of time on certain days and other days, so you have all your meetings in one day.

So it sounds a bit like Paul Graham's Maker’s Schedule versus Manager's Schedule instead of having the Zebra calendar. You have that space, especially because we now know that sitting down for an hour to try and do an important task is like not getting through it. 

Sometimes you need some time to mess around, get inspired, and then have the time to actually dive in and reflect on that, maybe while you walk out. So having those big blocks of time obviously matters a lot.

You talked about modern work. So what does modern work for you look like? And again, what are some actionable things that people can apply to how they work? Maybe in very antiquated ways.

Annie Dean: Yeah, it's interesting. Maybe the best way to describe modern work is to talk about where we've been and where we're going. So I think that the way that we developed offices and how work gets done is in this context where everyone is sitting near each other and working together, right?

Daan van Rossum: Which hasn't happened for about three decades, I think, but yeah. 

Annie Dean: Exactly. I think that Mad Men is funny; did you ever watch Mad Men?

Daan van Rossum: Of course. Yeah.

Annie Dean: Okay. So I feel like Mad Men is like a love letter to the office. It's this world where everyone sits in this incredibly beautifully designed space, and they're all just creative all the time. When they're not having free martini lunches, that kind of way of working peaked. Maybe in the post-FANG boom, it started to hit its peak decline. It started to head into its decline around 2018.

I think that in those models, relationships were the key to sharing information and getting work done. So you had to use the game of telephone in order to get information, relationships, and culture around a company. And what I mean by the game of telephone is that Annie meets with Daan, Daan meets with Molly, Molly meets with Christine, and Annie meets with Daan, Molly, and Christine. You know what I mean. It's all using these groups of people to get bits and pieces of information throughout the company.

And the culture is defined by its physical space. You walk into the space, you know how you're intended to feel, etc. I guess there's one more factor there. Everything's in bits and pieces. So some information is in my head, some information is in Daan's head, some of it is in my notepad, and some of it is in an email thread that's not that easy to navigate. It's all quite fragmented.

But it's doable because everyone's proximate. So the error rate gets managed because I bump into you in the cafeteria, and I completely don't understand that email thread. And you're like, Oh, it's just about these three things. So in the new world, none of those things are important to optimize for. They're actually just wasteful.

We don't actually need to manage information through a game of telephone anymore. We can just build an async system where everyone can access it at any time. And we can use AI to make that async system incredibly easy to access, manage, and utilize.

So instead of information being lost in my head, in an email thread, or in a couple of Slack messages, that all exists in a digital ecosystem. Instead of just speaking out loud and sharing information, we actually use a much more rigorous form of thinking, which is writing, and we put things into writing, get them onto the system, and then we can ask AI, What does Daan think about this current proposal? Who does Daan report to? Are there any other teams at this company that are doing this kind of project? Can you show me the three documents that I reviewed this week with Daan? And can you summarize them for me?

And that is much more effective because I can take those actions instantly when I want to take them and get results in a matter of seconds. Each of those things that I described would have taken hours because I'd have to connect with Daan. I'd have to understand what he thinks. I'd have to synthesize it. We'd have to have an exchange. That would happen in between the small amounts of time I have between meetings.

So when you move into this new modern work system where information is readily accessible and work is managed by technology instead of by relationships, you start to get to a place where you get a lot more time. And so you can move out of this work-about-work model that we're stuck in today. And into a place where we can start doing core work again, because I think what we're learning is that this system, the relationship system, is so hard to manage. It's so manual that we're spending all of our time on work about work, and we actually don't have any time for core work.

Now, I also don't want to freak people out and say that modern work is not about relationships. Of course, it's about relationships. It's about deep, real relationships that support solving problems, deep connection, etc., because we actually have time to invest in those relationships because we're not just spending messages, trying to get across the finish line, or organizing to move something forward.

So it should be a healthier system now. We could totally fail at it and just let AI make it messier and harder and try to participate in a thousand meetings instead of 10 meetings by watching recaps and those kinds of things, but that's where clarity is an important human task. We have to know what matters. We have to orient ourselves around what matters, and if we can, then I think that there's a really positive shift we're heading into.

Daan van Rossum: Yeah, and I think you've also said that people need to become creators again. They need to get out of being administrators; I'm just doing the whole day; I'm just answering emails, being in meetings, and reporting to people.

All that stuff can be done by AI, and I think that's the main role that should be the main role for AI—that it actually takes over all the dread, all the toil, and all the stuff that we don't like doing so that we can focus on the things that matter. 

So a note on clarity, because I've heard it a few times. Clearly, we have to ask the question right. Why is it so difficult for people at work to be clear on what they need to do? And I know you run some small hack to make that happen by having a chat bot. I think you say to people every day, What is your key priority for today? Maybe you can do some checking, but why is it generally that people are so bad at knowing what's important because you think you would know what your job is?

Annie Dean: Yeah, I'm sure Molly would have a deep psychological response as to why humans are bad at knowing what's important, so I'll leave that to the scientists. But as we think in practice about work, one, I think we're out of practice because our attention has been so taken from us and we're so habituated to that reality that it's hard for us to step above the noise and really declare what's important.

For instance, you think about work calendars and the fact that most people show up to their work day and let their calendar tell them what's important. The calendar is a boss. And I always joke about this, but I'm like, Your calendar is not in control of you. And I think it can be unsettling for people because I do work very differently. I wake up, and I'm thinking about how much time I have left in the unit of the week. What have I left unfinished? What do I need to deeply invest in? And my request for people's time can be disrupted because I'm willing to say my calendar is irrelevant; I'm focused on what really matters, and that takes some adjusting, by the way, because it requires flexibility. So, I think we're out of practice.

Two, I think you really need sponsorship from your manager, and you can build that both tops down and bottoms up. You can really advocate with your manager to say, I really want to know what matters and what's important and a priority.

As a leader, I think you also have to be willing to place your bets and say, I have a good planning process. I'm looking at the right information. I'm good at making a call on what matters to my business unit, and I've socialized that to make sure that I have air cover to know what's important without letting go of the ability to be flexible and to react to the shifting environment that we're in as time passes.

Daan van Rossum: But of course, the reality is that those managers also just wake up on a Monday morning, completely overwhelmed by the work that they need to be doing and by the meetings that they have to take on that day.

How do you get out of that? How do you create time to even set priorities? I think that's going to be a very practical thing that people run into. 

Annie Dean: It's a very self-fulfilling prophecy. For instance, I just elevated somebody into a new role, and she has double the responsibilities that she had before. And I was like, The thing I'm most worried about for you is that you're not going to have enough time to engage in what matters because you're going to still be stuck in the details of the job that you were in before.

And so I was like, the most important thing for you to do is to just immediately clear 35–40% of your calendar. I know that's going to be really painful because people are going to want your attention on things, but you need to sit in this and get clear and really be able to elevate your perspective, absorb information, and see where people's requests are coming at you in an urgent way that you can't avoid.

Because once you introduce scarcity, you can only utilize a certain amount of time to meet those requests. It's a natural prioritization method. And then, when you have free time, you can fill that time more thoughtfully and with greater perspective.

So, it's hard; it's difficult. It's like anything else—building a new skill. But I think by immediately creating more time, it gives you the reduced temperature, perspective, and awareness to say, I'm a little bit closer. At least I can facilitate the conversations about what matters differently now.

Daan van Rossum: It's really a new skill and also a new habit. So when I do coaching on how to use AI and be more productive, and again, productive in the sense of meaningful productivity, what matters most to me is what I want to get done, and therefore you need to know what your company wants to get done and what your team needs to get done. 

I always quote Seneca, 2000 years ago; nothing new about it. But we seem to be very loose with our time and very frugal with things like money or property. I think that's his quote. We're very frugal with money and property, but we just give it to anyone and anything. Time is actually the only truly precious thing we have. It's actually the only thing that scares us in our lives.

So maybe now, switching back to 2000 years ago, let's not do that. But let's switch back to today. 

Whether within Team Anywhere, within Atlassian, or just more broadly, what are some ways that AI can help us win back that time so that we have that space? I think a lot of people feel suffocated when they look at that calendar, when they think about their to-do list, which I think you always say. It's really what other people want you to do, not what you want to do. People have the feeling of being suffocated. So how do we maybe use something like AI to free up more time? 

Annie Dean: I think first of all, what we're finding is that teams are not fully adopting AI yet. They're not crossing the chasm, and yet those teams that do are doing almost twice as well. And so we know that AI is a really powerful tool for unlocking teams. I can say that there are a couple of ways that I think I'm using it today that are good and experimental.

If you're listening to this and you haven't experimented with AI yet, just literally go ask ChatGPT something, and you'll see how easy and intuitive it is and realize that it's not. There have been other technological changes that have been more difficult to understand. But actually, what's so crazy about AI is that it's incredibly intuitive for us as humans because it speaks to us like a human. So it's really easy to use once you force yourself to do it.

Daan van Rossum: It's the most human technology ever invented. In fact, I think Ethan Mollick and others would say it is not technology. It's a new coworker. It's a new colleague. So you just speak to it as if you were talking to someone. 

Annie Dean: And I think that's really important to understand, and as we have developed tools at Atlassian, we are actually calling it an AI teammate called Rovo, and it's incredible, which I'll talk about in a second.

How does AI help? One, if you haven't made a Loom yet, just go make one because it's one of the most effective ways to replace meetings, and we found that it's been incredibly effective to keep teams connected and aware of what's important, as well as to just extend a birthday message or something else that might be personal but in a difficult time zone.

With Loom, it used to be that you would have to make the video, and then you would title it and add a caption. Now you just record the video, and it's titled, it's auto-captioned, it's divided into chapters, and it's summarized. That's a way that AI can really help us, because it's making the technologies that serve us so much more intuitive and better. And it enables us to just take the one action that matters. It's insane.

Daan van Rossum: That's magic. When I send that Loom and it immediately has the title and I don't have to think about that again, that just cuts that workflow in half. 

Annie Dean: In half, and that's exactly how the great technologists right now are figuring out exactly how to optimize their products for that. I was in a conversation not long ago with some of our leadership, and they were saying AI is a great technology for incumbents because who's going to win the AI war is going to be these AI startups, or is it going to be incumbents that integrate AI? But the reality is that leading companies today are implementing AI into their product stacks and making their products do much better. We, as consumers, are really going to reap those benefits. So that's one.

Two is that we're regularly using Rovo, which I was describing as our AI teammate at Atlassian. It's awesome. And some of the ways that I've been doing it lately that I referenced earlier are things like never finding anything. You know what I mean. I don't know if you guys are like this, but unless I've made a commitment to myself because I am an insane organizer, you should see what my fridge looks like. I love minimalism and clarity. Every drawer in my house is perfection.

Daan van Rossum: The least surprising thing you've said in the last 30 minutes, but yes.

Annie Dean: But when it comes to the digital ecosystem, I made a decision back in 2021 that I would not manage it because it's not worth my time. So often, that puts me in a position where I can't find something, and that's super frustrating to somebody who wants a perfect drawer system in life.

AI solved that problem. So now I go and ask Rovo, Hey, I was working on a document with my colleague, Zach. Can you pass it this way and summarize it? And it is incredibly effective because it takes 10 seconds to be able to do that. My old way would be to go ask Zach because this is like the most important document that he's working on, and he would be able to send it to me instantly, but then, like, hours go by, and is he going to send it to me at that window where I have the time to look at it?

That kind of thing. So that's one way, and that's quite administrative, and I think people think that AI is very administrative, and it is, but it is also a collaborator. So another way that we recently used it is that I was in one of those long sessions, and I was with a number of my teammates, and we were actually really struggling to try to explain what async is because I think it's really important for people to understand that async and AI are the backbone of how work happens, not the office, and all these things are a little complex. What does async mean?

It really means something way bigger than that word, and we asked Rovo to explain it to us like a kindergartener, and it came up with such great intuitive language that we could actually utilize, and frankly, we could have spent three more hours talking to each other and we would not have gotten to that quality of answer.

Acting like Rovo or an AI assistant is a collaborator can yield really great results. And I think one of the things that we've noticed as we study the adoption of AI is that there's a great team member on our team named Ben who's so curious about AI and how teams will adopt it, and we've found that people are very transactional with AI. They ask it one question, and then they're done. But they need to remember that it's a conversation that's being had, and that gets to a different level of impact.

Daan van Rossum: Absolutely. And it's not just one colleague; it's a team of colleagues. I always tell everyone it's... 

Annie Dean: It is the smartest colleague that ever existed, and then a final way that we're using it is to always say fancy words to each other, like with a wink. But there's this concept of agentic workflows. And what that really means is that you teach an AI to be really smart at a thing. And it does that thing for you. And I think when we think about programming something, we think of it as really complex. We need an engineer. We need a designer. We need a front-end designer. We need somebody to think about UX.

Well, in this new world, all you do is pull up the AI and say, Hey, I'd like to build an agent, basically. And I'd like you to be an expert in this topic, and as you discuss it with me, I'd like you to prefer all of the things that are basically in this folder that I'm about to share with you. So can you read everything in this folder and become an expert on it? And then I'm going to ask you questions about it. That's all that it takes to program this expertise.

So it's natural language programming. And I don't think that people understand how simple that is yet, and once they do, it's just going to really explode.

Daan van Rossum: Yeah. It's always the mind-blowing thing of, Wait, this is what AI is. This is how simple it is. This is how human it is. It's just so mysterious for people, but actually it is very simple. So I know we're totally out of time. So I'll just ask the very last question very quickly—two seconds. Your big wish for the future of work.

Annie Dean: I love that question. My big wish for the future of work is that everyone gets to live a little bit more, be healthy, and reconnect with their creativity. I think that's what the world needs right now. Really creative problem solvers with the capacity to go do really hard things, and I really hope that people who have been left out of the workforce to date are the people running our companies in 10 years because we've introduced these models. I hope that brings us new perspectives. So let's see.

Daan van Rossum: That sounds amazing. Thanks so much for being on, and thanks so much for closing this season with us. 

Annie Dean: Thank you, Daan.