#82: The Intersection of AI and Agile with Emilia Breton

Join Brian and Emilia Breton as they explore AI and its practical applications in Agile, from enhancing productivity to coaching. Don't miss their insights on AI tools shaping the future of Agile.

Overview

Today, Brian sits down with Emilia Breton to unravel the dynamic intersection of Artificial Intelligence (AI) and Agile.

Brian and Emilia share their experiences and experiments with AI tools, revealing how they leverage these technologies to enhance productivity and decision-making and amplify human capabilities.

Listen in to learn more about the evolving landscape where AI and Agile converge to shape the future of work.

Listen Now to Discover:

[01:25] - Brian welcomes Emilia Breton to the show to talk about the intersection of AI and Agile, focusing on using AI to enhance human connections.
[03:04] - Emilia shares that it's about using AI to accentuate our humanity and create space for us to connect, observe, and inspire.
[05:15] - Emilia discusses the questions about copyright for AI-produced content, such as code and why It's important to be able to trace where AI-derived information comes from.
[06:02] - Brian reflects on the rapid evolution of mass-consumable AI and its transformative impact over the past year.
[06:39] - Emilia underscores the importance of visibility in AI outputs, and the need to cross-verify AI-generated information with human expertise.
[08:41] - Brian introduces the concept of hallucination in AI, emphasizing that AI can't think or reason, and it may generate false information to please users.
[10:29] - The importance and irreplaceable qualities of human competence.
[11:30] - How tools like Lucidspark can help with ideas during product brainstorms or retrospectives. Otter.ai and Spinach.io can automate tasks like taking meeting notes and updating Jira, saving time for more important work.
[14:16] - Brian introduces Rewind.ai, a tool that records computer activities for later recall, explaining its potential benefits and privacy considerations.
[16:10] - The Agile Mentors Podcast is brought to you by Mountain Goat Software and today’s episode is brought to you by Mountain Goat Software's Certified ScrumMaster Class a two-day class covering the fundamental principles of scrum as well as detail about the different roles, meetings, and artifacts. For more information click on the Mountain Goat Software Certified Scrum and Agile Training Schedule.
[17:01] - Emilia explores the use of AI to spark inspiration, and shares generative AI art programs DALL·E or Midjourney.
[20:06] - Emilia discusses using Grammarly and AI as a partner in content creation, to iterate prompts to achieve the desired tone and style.
[21:10] - Discussing Google's new multimodal Gemini models, which can translate speech, text, video, and images.
[22:07] - NotebookLM is designed for researchers to organize and refer to research papers and articles. Brian shares his experience with this tool.
[22:48] - The speakers discuss using AI for data analysis to interpret large datasets, capture notes, brainstorm ideas, and facilitate retrospectives to enhance Agile practices.
[25:08] - Brian and Emilia discuss how AI can be a valuable tool in coaching and can assist in facilitating sessions.
[28:08] - What lies ahead with AI?
[29:24] - Brian sends a huge thank you to Emilia for being on the show. If you found this episode useful, please share this episode with others. We’d love your feedback and suggestions for future episodes. You can email us at [email protected]. And don’t forget to subscribe to the Agile Mentors Podcast on Apple Podcasts so you never miss an episode.
[30:09] - If this topic was impactful to you and you want to continue the discussion, join the Agile Mentors Community where we have a topic discussion for each podcast episode. You can get a free year-long membership in the community just by taking any class with Mountain Goat Software.

References and resources mentioned in the show:

Emilia Breton
Lucidspark
Otter.ai
Spinach.io
Rewind.ai 
DALL·E
Midjourney
NotebookLM
Subscribe to the Agile Mentors Podcast on Apple Podcasts
Certified ScrumMaster Training and Scrum Certification
Certified Scrum Product Owner Training
Advanced Certified Scrum Product Owner®
Advanced Certified ScrumMaster®
Mountain Goat Software Certified Scrum and Agile Training Schedule
Join the Agile Mentors Community

Want to get involved?

This show is designed for you, and we’d love your input.

• Enjoyed what you heard today? Please leave a rating and a review. It really helps, and we read every single one.
• Got an Agile subject you’d like us to discuss or a question that needs an answer? Share your thoughts with us at [email protected]

This episode’s presenters are:

Brian Milner is SVP of coaching and training at Mountain Goat Software. He's passionate about making a difference in people's day-to-day work, influenced by his own experience of transitioning to Scrum and seeing improvements in work/life balance, honesty, respect, and the quality of work.

Emilia Breton is an Agile wizard with over two decades of experience, who effortlessly navigates the realms of startups and global corporations. Specializing in guiding both scrappy ventures and colossal entities, she brings innovative approaches to software development and team building. Emilia's commitment to injecting playfulness ensures a dynamic and creative touch to Agile practices, making her the go-to coach for those ready to elevate their software development game.

Auto-generated Transcript:

Brian (00:00)
Welcome in Agile Mentors. We are back for another episode of the Agile Mentors podcast. I'm with you as always Brian Milner, and today I have a very special good friend of mine who's come on with us. Miss Amelia Britton is on with us. Welcome in Amelia.

Emilia (00:15)
Thanks, Brian. It's great to be here.

Brian (00:17)
So glad we've finally been able to do this. I've been trying to arrange our schedules to do this for a while. So it's hard to really pin Amelia down on one thing that I would say, here's who she is, because she's a lot of things. I think coach, if I just said coach, I think that would be a very good term, because she's a very, very good coach. And... She's credentialed with the Scrum Alliance in that area. She's got the highest credential with IC Agile in this area. She's an IC Agile expert. She's worked the gamut. She's worked with startups. She's worked in very, very large companies. And we got to see each other in person at Agile 2023 last year. And really, that was the first time I think I've seen you in person in a while. And it just was really fun to see you and get to catch up there. So we thought we'd get Amelia on. And she's been doing some talks recently along a certain topic that I know people are really, really interested in. But it may not be exactly where you think we're going with this. But really it's talking about AI and Agile as it plays into AI. But like I said, it's not going to be, I think, where most people think we're going to go with this. So I'm talking too much. Emilio, why don't you take it from here? If it's not going to be where most people think we're going to go, how would you sum it up where we're going with this?

Emilia (01:49)
I think really it's about using AI to accentuate our humanity and create that space for us to do what we do best as humans, which is connect and observe and inspire.

Brian (01:54)
Mm, I love that. That's awesome. Yeah, I love that. So just to lay the foundation here a little bit, neither Amelia or myself are AI engineers. Neither of us work for open AI or anything like that. We are not in the thick of that AI world. We're consumers. We're consumers of it. And we're tech people. So We've spent our lives in and around software development teams. And that gives us a little bit of a unique perspective. I think that combined with the combining Agile with AI, I think, also is the other unique kind of perspective we're trying to bring to this. So let's kind of dive into this a little bit. When we start talking about AI and Agile, What are the kind of considerations you think are the most important things we should be dealing with when we start to delve into this area?

Emilia (03:07)
Yeah, I think one of the really first important things to know is to really think a little bit about the sort of ethical considerations when you're starting to use AI. You know, there's there are many privacy concerns. You know who owns the thing that you've created. Do you own it. Does it belong to everybody. The ownership of the things that this was trained on and knowing what it was trained on and where the data comes from. How do you keep your data safe and secure? Anything, you know, it all depends on what are the rules of the thing that you're putting your data into. Sometimes you're putting data in and you retain it as exclusively yours. And sometimes you're giving it to the model. So you want to be really careful in knowing what you're doing, knowing what the model is, and then making your own sort of ethical determinations around that. and what you want to use.

Brian (04:03)
Yeah, there's a, you know, I dabble in music sometimes and there's some interesting channels I followed on music with AI. And, you know, there's kind of a, there's one gentleman in particular who talks about copyright and music and AI. And one of the things that he proposes is that if it's AI produced, that he thinks there should not be a copyright for the content because it's not created by a human, it's created by a machine. And I found it to be kind of an interesting take on it. You know, and I wondered how that would play in our space. You know, if a machine writes code, is it private?

Emilia (04:41)
Yeah. And there's a lot of questions that kind of come around that. And right now, it looks like if you're following any of the big legal cases that have happened, that it may actually lean to, if you created it with the AI, you might not own it. So it's a consideration if you're starting to use any of these tools. Another one that's really big for me is explainability. So being able to tell.

Brian (05:05)
Okay.

Emilia (05:07)
where the AI got this information. Different models, different applications, you are better or worse at this. Certainly a lot have gotten better over the last few months. I mean, we're only really a year into this mass consumable AI world. I mean, certainly AI has been around since forever and big blue, but the mass consumption of it being available to you and I to just...

Brian (05:24)
Right.

Emilia (05:34)
go on the internet and use really came with chat GPT and those a year ago this week. So it's a new field. And we've gotten this much in a year, which is just, you know, we're in technology, we're in one of these growth curves. So everything's changing. But explainability like Google's model probably does it the best. They give you really great links. Others, you just don't know.

Brian (05:41)
That's amazing, isn't it? Yeah.

Emilia (05:58)
And there's a lot of push within the AI ethics community to really make that visible and be able to see. Because AI actually isn't competent, right? It doesn't actually know the things. In pretending, it is performing of being intelligent and competent. And so it's also important to keep that in mind. And that's one of the powers that we have as humans, is intelligence.

Brian (06:09)
Yeah.

Emilia (06:26)
We actually do have that competence to be able to check the AI, to be able to figure out, you know, is what it's saying true or not. There was an instance I was creating, I had a list of books and I said, OK, great list of books. Tell me more. Give me some more answers that are like this. Right. And one of the books it came up with, it was perfect. It was talking about

Brian (06:43)
Yeah.

Emilia (06:53)
Agile in Enterprises versus Startups. This was like exactly the book I wanted to read. I was so excited about it and it told me that it had been written. I think it was Diana Larsen and someone else. I don't remember who the other person was, but I was really excited because these were two people who I was like, oh, yes, I know they collaborated on a book. This is gonna be fabulous.

Brian (07:13)
Oh wow. Yeah.

Emilia (07:21)
Um, and I went to Diana's site and it wasn't there. It was like interesting and then Amazon and wasn't there. Um, and I went to the other individual site and it wasn't there. And I went to Lean Pub because you know, a lot of times we as Agilist will start something on Lean Pub. Um, and it wasn't there. Um, and I DMed Diana. I was like, I'm so excited. Tell me more. She's like, uh, not a book.

Brian (07:25)
Ha ha ha. No idea what you're talking about.

Emilia (07:50)
like, well guess what book you should write? Because it would determine like this is the book that you want to read and it fits with these other books. So interesting and scary but I totally hallucinated this thing existing. Had an ISBN number, had the authors, had a publish date, had a publisher. All the things that looked like a proper citation.

Brian (07:53)
Right. Yeah, that's an important thing for people to know. If you haven't really gotten too far into it or if you really kind of stayed out of it, the term hallucination is a term you'll become familiar with at some point because like Amelia said, it doesn't think, right? It can't think, it can't reason, but it is programmed to try to please you. It is programmed to try to give you what you want. And if it doesn't have the data, Sometimes it does this thing called hallucinations where it'll make it up. It'll kind of create it based on the data that already has. Yeah, I remember I've tried to push it in certain ways. I like to do a lot of just testing the boundaries. So I know from early on to, you know, for the past year, there have been times when I've done things with things like Chat GPT, where I've tried to push it and say, what's the answer that you have for this or... You know, and one of the things with Chat GPT is it's not, or at least for a period of time, I guess now it is kind of connected, but for a period of time, it was not directly connected to live data. So there's no way you could have it search the internet and find something out. Now there are ways that you can have it do that. But I remember early on trying to get it to tell me the score of a game. And it was a professional sports game. And it said, oh, yeah, here's the score. Here's what happened. These are the stats from the game. No, that game hasn't happened yet. It's this date, and that's not happening till. And you'll get that typical response. I apologize. You are correct. I did not. I apologize for giving you incorrect data. Right.

Emilia (09:55)
Yeah, I've only been trained till September 2021.

Brian (09:58)
Right, right, right. So that's something to just be familiar with is, like you said, it can't reason. It's really approximating what it's doing to you based on all the huge volume of data that it has and trying to craft a response that is similar to the data that it's already been fed. So I'm probably telling people things they already know, but just in case you didn't know that, wanted to state that.

Emilia (10:23)
But it's super important to always keep that in your head. And this is one of the places where AI really can't replace us. We are truly competent. We are truly creative. We are truly innovative. And that's the place where we as humans really stand out. And I've been using AI for a bunch of things and experimenting with at my work. Fortunately, I work for a startup, so things are much easier to get through. It's a matter of having a conversation. having one of our wonderful legal team take a look at things and then give me the go ahead. I don't have a lot of red tape that I'm sure some of your folks who are in larger enterprises may have. But I've had the opportunity to play with a bunch of things that have been really helpful. I sort of, there's like two sort of categories of things I've been looking at. The first one is stuff that automates the non-promotable work. So.

Brian (11:15)
Yeah.

Emilia (11:16)
All of the little things that we have to do that take a lot of time, but nobody's gonna remember you for. Nobody is going to remember you for taking wonderful notes of that meeting. Nobody's gonna remember how accurate you were at updating Jira. People are gonna remember the impacts that you made on them and their teams. And that's the work I wanna spend my time doing and not spend as much time on the other stuff.

Brian (11:43)
Right.

Emilia (11:45)
So I've played with a bunch of different apps. I don't know if it's okay to talk about specific ones. Yeah.

Brian (11:48)
No, yeah, please go ahead. Yeah. I'm sure people want to know some, want some tips. Yeah.

Emilia (11:53)
Yeah, so like one I love is Lucid Spark. So they are one of the mass of collaborative whiteboards and they have some really beautiful AI features. They're just ubiquitous built into their product that help when I'm facilitating. So I can have it take our massive wonderful, creative, messy ideas that we've all come up with when we're like doing a product brainstorm. And then... sort them all up so that we can really quickly like in moments as a team look at okay here's some different ways that this might be sorted where do we want to go next what do we want to do next um and that's just a super useful piece that they have and then say i'm done with that or i'm done with a retrospective to be able to get a okay let me select all the stuff and then give me a summary of what are all the things that happened

Brian (12:42)
Yeah.

Emilia (12:50)
And then I edit it because we're humans and we need that part, but it saves a lot of time. These are pieces that are just, it's using the technology to give us time to have better conversations, to have to work better as a team. It's just taking that out. So that's one that I love. There's another.

Brian (12:56)
Yeah. Yeah, I think of that, you know, I've tried to say to some people that it's more on the level of like a spell check, you know, like I don't I don't look at it like, you know, like I said, it's not it's not creating its own ideas. It can help you brainstorm, it can help you do things, but it does still have that human element that it requires to refine and to push back a little bit and say, yeah, but what if we went this way? So it's good if you look at it as this is another tool, this is a tool that can help me do things, yeah.

Emilia (13:43)
snack them. And there are a ton of tools that do that. You know, there's some great tools that will take your meeting notes for you. Otter.ai, Spinach.io are two that I've used that are pretty good. You know, Spinach has the plus of, it'll go update my JIRA for me, which of course we love. But they're things that are, they just take all of the bits and pieces and they do them for you. so that you can spend the time on the important stuff. And that's one of the things I'm loving with a lot of these nutrients.

Brian (14:14)
That's awesome. Yeah, there's one that I've come across as well. I'll just throw out there. And we're not plugging anything, trust me. We're just sharing our own experience. But there's one that I have used a little bit often on an experiment with, it's called Rewind. And yeah, now some people might think it's a little creepy. So let me explain and then you can decide if that's something you wanna experiment with or not. But Rewind basically kind of records everything on your computer.

Emilia (14:21)
No. Yeah, that's a pretty good one as well.

Brian (14:44)
I mean, you can tell it what not to record. You can say, don't do this. And you can set it up so that it will ask you any meetings. Should I record this one or not? But basically, it's sort of there in the background. And then you can call it up and say, I saw a website that was about this, or I saw something this week, or there was a meeting I was in where we talked about this topic. Can you recall that for me? And it'll say, yeah, here's the meeting, and here's the transcript of that part of your meeting where you talked about this thing. So I've heard people say it's sort of like my extended brain. And it does, I mean, my understanding is it's just local. It's not putting that out into the. world anywhere, it's not saving your data anywhere else, and you can control how much data it saves and everything else. But I understand that might have some privacy concerns, especially depending on someone's industry. If you're in a very highly secure industry, that might be a huge warning flag. So it's to each their own, I get it. But that's kind of where it's at, that boundary of let's see if it can do this. Oh yeah, it's actually. It actually can do this and it can do this pretty well. Yeah.

Emilia (16:04)
Yeah, and there's also like that's consent at that point becomes really important. Right, you're putting on your choosing, you're giving your consent to this thing. It's something you know, you wouldn't want somebody to impose upon you. That's when it goes from the useful tool into like creepy dystopian world, right? It's all about the consent and all about knowing and traceability and that kind of thing. And I think, you know, when you talk about some of the

Brian (16:23)
Right.

Emilia (16:32)
ethics folks, they are worried about crossing that line. It's just a matter of being aware, I think, at this point. And the other play way I've been using a lot of AI is for really working and helping it to spark and inspire stuff. And I've been doing a lot of that with using it to create metaphors. So.

Brian (16:39)
Yeah. Yeah.

Emilia (16:54)
One of my favorite things I've used it both in one-on-one coaching and then also in like team retrospectives is using some of the generative AI. I've used some of the generative AI music, some of your generative AI art programs like Dolly or Mid Journey to take what somebody is feeling, thinking, seeing about the situation, the sprint, and then having them write the prompt. to create that piece of generated art that they can then share because it takes us beyond our words as humans. And into that so that you can see visually or hear what I'm feeling and it creates this lovely emotional tone that lets us go into different places and ask different questions of each other in say a team environment or of an individual ourselves in a coaching kind of environment. It's really powerful because it's taking, you know, before we would say, okay describe it, tell me more about that. But to be able to show it, put it on the table, and then for that individual to look at it, or to listen to it and pick a part and tell me about this part, tell me about that, how does this connect? It just creates all of these other areas that we can explore. And that's the stuff like we're humans when we really excel.

Brian (18:16)
Yeah. Yeah. Yeah, that's a great point. And it's, you know, when we forget that, when we forget that and try to lean too heavily on it, then it's painfully obvious. It becomes really easy to spot and you're like, this is not human. This is not, this has no soul to it. This has no, you know, creativity or imagination. It's just spouting back stuff that I've seen and heard before.

Emilia (18:46)
million.

Brian (18:46)
But yeah, I think you're right. You know, if it can inspire me a little bit and say, yeah, I'm looking for something like this, give me some examples. You may get 10 examples back and then think, no, none of those are exactly right. But I like kind of how number seven talks about this. What if you gave me several examples like number seven, but tweak it this way? So that's the kind of. For me, that's the kind of back and forth that it takes to get something usable. It's just not write a book, here's the book, you know?

Emilia (19:20)
Yeah, I think one of the things that really we as Agilent have the power when we're using some of those AI things like chat GPT to generate content is the iterations that we're so good at. You know, I've done some work with like, okay, I'm going to use GPT. I'm going to use one of the large language models. Let me to get me started on writing a large piece of content. And then you iterate, you're like, no, more this, less this. What I actually like in that is less than starting the content and more, here's my piece of not very well-written content with all of my ideas and all these things. OK, so help me turn this into a tone that is more ABC. I love Grammarly for this. But give me a more playful tone. Give me a more.

Brian (19:58)
Thanks for watching. Right. Yeah.

Emilia (20:14)
business tone and you can take the same piece of content that you've written and then tailor it to different situations. Well it's still you. And iterating on that prompt like, you know, less marketing jargon, more, you know, and really till you get that piece. Kind of acts as a partner to help you inspect and adapt. To get it just the way you want it.

Brian (20:24)
Yeah. Ha ha ha. That's awesome. Yeah, and I like the idea of thinking of it kind of like a translator. You know, like when you talked about the images and stuff, that's what was going through my head is, you know, there's some people who do really well at creating images and drawing and digital artwork and everything. And there's other people who don't really know that language, don't know how to do that as well. But if we can have something that I can describe it to and it can create it for me, it's just a translation. It's taking my words, it's taking what my idea was and showing me a version of that. I can tweak it, I can push it one way or the other, but it's just giving a vocabulary, if you will, to my idea.

Emilia (21:26)
Exactly. And I think these are the things that are only going to get better. You know, we just recently had the release of Google's new Gemini models, which are multimodal, which is a whole other realm. So how can I take and translate, you know, speech and text and video and still images?

Brian (21:37)
Yeah. Ha ha ha.

Emilia (21:51)
and bring them all together into something. I'm really excited to play with them as they start hitting because I think that's the thing that's really going to unlock our power to do some really neat things together.

Brian (21:56)
Yeah. Yeah, I agree. By the way, there's a tool I'll mention as well that I've played with recently that I'll give a shout out to. Because for some of the stuff I've been trying to do, gather research and draw some conclusions from it and everything else, it's been a really helpful tool for that. Google has this tool out that's called Notebook LM. It was designed specifically for researchers. It allows you to, it's in the same mode or realm as sort of a custom GPT kind of thing, but you feed it research papers, articles, you feed it things that you've been using as a basis, and then it can help you. to refer back to the right thing and say, show me all the research about this, and it'll bring it up for you. So it's sort of just like a research assistant that you can use and you can create different little notebooks for different topics and other things. So I found that to be really, really useful. And that's kind of one of the newer tools that's out there in their experiments as well.

Emilia (23:08)
Yeah, I've been experimenting with OpenAI's data analysis and feeding it. So I have a set of surveys that I use with teams to really kind of look at, you know, where's your agile fluency? What are some of the markers of this team on their journey and different things like product, et cetera? And then feeding it these things and having it, you know, tell me more, you know, give me some analysis on this.

Brian (23:13)
Hmm, wow.

Emilia (23:33)
And I found it one that I had done previously. Usually when I take in a bunch of them, it takes me about a week to compile things, look at standard deviations, pull some things out about what I'm seeing as patterns. And it was able to produce the stuff that usually takes, again, I'm both human, around a week. I was able to do it. I mean, it did stop and start and feed and ask questions. And so there's tweaking. but probably within about 10 minutes on not terribly well clean formatted data. And its conclusions matched very closely with what I had done as far as analyzing the data. Now, of course, we take that and then based on that data analysis, what does that really mean? That parts us as humans. But it's a huge time saver when you're looking at large pieces of data.

Brian (24:19)
Right. Yeah, it's fascinating the kind of tools that are available to us, like you said. So this is great. I just want to reiterate a few things here. If you're... Working on Scrum teams, and you haven't considered doing something like this in the past, there are tools that could help you capture notes from maybe a sprint planning session or a sprint review. There are tools where you can help brainstorm new ideas for a retrospective or even take action items from that retrospective when you're done. So. This is what I'm saying, this is what I meant at the beginning. We're not going into the typical kind of areas with this. There are small little things that you can use this for today that I think can really give you a boost and enhance your practice.

Emilia (25:16)
Exactly and give your team the time for the stuff that your team is really great at It gives you the time more time to collaborate more time to have the conversations about how might we or what might we? Where we are wonderful Nope like I said before nobody's gonna remember you because you took great notes

Brian (25:21)
Yeah, that's great. Right. What about, and I know we're pushing up against our time, we'll try to wrap things up here, but what about because you have a lot of experience in the coaching realm, what about tools in that kind of aspect? How have these tools helped in sort of coaching practices?

Emilia (25:54)
Yeah, so one, obviously the metaphor I spoke about earlier, helping inspire those ideas and let us ask deeper questions, kind of, it's really great to put, say, a picture out and be able to tell me about this, tell me what's the significance of this, what do you see, what do you, or with team, what do you notice about this pattern of pictures? That's one I've used. Another one is I've used it just for my own sort of self-improvement.

Brian (26:21)
Hmm.

Emilia (26:22)
Um, I have a custom GPT that I trained and I fed it, you know, a bunch of stuff around ICF coaching ethics and how to be a great coach as far as those things. Um, and then I fed it some other, um, agile resources like the agile coaching growth wheel, um, and I've used it just to sort of have it ask me questions as I start exploring where do I want to go next. What do I want to do next? And I certainly have deeper more meaningful conversations with my human coaches Who coach me because I believe everybody needs a coach But they help me sort of sort my ideas and think about different areas and different things That I might not have thought of They're certainly not actually coaching me, but it does do pretty well at you know

Brian (26:57)
Sure. Ha ha.

Emilia (27:15)
pulling out those pieces so that I can think about what I need to think about and get a little bit out of my head. And it's useful for that.

Brian (27:21)
Yeah, that's awesome. Yeah, I haven't even thought of using it for something like that, but that makes perfect sense. Again, it's just a tool. It's something that you can use to give you a push maybe in one direction or the other. But at the end of the day, it's still on you. It's still on you to actually do the work. That's awesome.

Emilia (27:37)
Exactly. Yeah. Another quick one I do is like, it's really good at taking all of your notes for a session and building a nice facilitation guide. It's easy to follow.

Brian (27:50)
I hadn't even thought of that. That's awesome. Yeah.

Emilia (27:52)
And like pieces like that are just like, just, it's like having, you know, it's like having an intern. Doesn't know anything. Who's brand new and fresh. But that fresh set of eyes can be useful.

Brian (27:59)
Yeah. Yeah. It's such an exciting time. I mean, I really, to me, it's just, like you said, it's only really been publicly around for about a year. And what we're talking about here is more these large language models. And we're not even really getting into machine learning and what might happen in this coming year with the combination of those two things.

Emilia (28:22)
Exactly.

Brian (28:28)
And then we're really cooking with gas, right? Then we're like, that's going to be amazing to see what kind of things come out of that combination. 2024, I think it's just going to be a really interesting year. You know, I know when that happened last year, I thought, you know, this feels a little bit like when the internet kind of launched. It's that kind of wild west kind of feel to things. And It feels a little like that's happening now, in this realm.

Emilia (28:56)
Yeah, I think it's going to be really interesting.

Brian (28:57)
All right, well, this has been. Yeah, I agree. Well, Amelia, this has been really interesting. I've had a really good time talking with you about this stuff. It's, you know, we could talk about this for a long, long time. And it might be interesting for us to check in maybe next year and say, look what's happened since the last time we talked. Here's where it was then and let's see where it is now. But thank you for proposing this and coming on and talking with us. And I really appreciate you sharing your research and learnings and insights on this recently.

Emilia (29:30)
Thank you.