What are flash teams — and what makes them such a dynamic model for rethinking how work gets done? Melissa Valentine, associate professor at Stanford University, discusses how AI and flash teams are redefining collaboration and productivity in the workplace. Reflecting on the power of AI, she states, "We are all so much more capable, augmented with these GenAI tools."
What are flash teams— and what makes them such a dynamic model for rethinking how work gets done? Melissa Valentine, associate professor at Stanford University, discusses how AI and flash teams are redefining collaboration and productivity in the workplace. Reflecting on the power of AI, she states, "We are all so much more capable, augmented with these GenAI tools."
Subscribe to The AI+HI Project to get the latest episodes, expert insights, and additional resources delivered straight to your inbox: https://shrm.co/voegyz
---
Explore SHRM’s all-new flagships. Content curated by experts. Created for you weekly. Each content journey features engaging podcasts, video, articles, and groundbreaking newsletters tailored to meet your unique needs in your organization and career. Learn More: https://shrm.co/coy63r
This episode is sponsored by Brightmine
[00:00:00] Navigating constant legal and regulatory changes is challenging. Brightmine HR and Compliance Center provides trusted, proactive updates, leading practices and tools to help you reduce risk and strengthen your HR strategies.
[00:00:14] Using AI assist simplifies research and decision making,
[00:00:17] empowering HR teams to act with confidence and reduce the time spent on manual searches. Request a quote at brightmine.com.
[00:00:26]
[00:00:32] Nichol: Welcome to the A IHI project. I'm Nichol Bradford SHRM's, executive and Residence for AI Plus. Hi. This week we're exploring human-centered AI and flexible team structures in the workplace. Our guest is Dr. Melissa Valentine, associate professor at Stanford University, co-director of the Center for Work Technology and Organization, and a senior fellow at.
[00:00:55] The Stanford Institute for Human-Centered AI or [00:01:00] the High, she's also the co-author of an upcoming book called Flash Teams. Melissa's research on Flash teams ai, augmented expertise, and organizational design is shaping the future of how we collaborate, lead and structure work.
[00:01:15] Melissa: Thanks. I'm such a big fan of the show.
[00:01:18] Nichol: Well, we're really thrilled to have you. I remember watching one of your. Lectures on video and thinking, oh my gosh, I really wanna talk to this extraordinary scientist. So thank you so
[00:01:33] Melissa: Of course.
[00:01:34] Nichol: Let's start with your journey in organizational science. What inspired you to focus on how technology and especially AI can reshape the way people work and collaborate?
[00:01:46] Why that topic?
[00:01:48] Melissa: Yeah, so I love that you noted that I start as an organizational scientist. 'cause I think you'll see that I bring that perspective to a lot of the discussion about ai. And actually the way that I got into [00:02:00] organization science, I started as a pre-med in college, but then I was like working in hospitals and I kept seeing all of these organizational problems and then it was like through that, that I actually learned that there was such a thing as organization science.
[00:02:12] So I got my PhD at Harvard Business School and that's where I learned about like organizational theory and I, I started to like really develop the point of view around organizations as this like tool for solving problems. So you want to provide healthcare for people, you wanna provide education, you want to like invent a new technology.
[00:02:30] Having an organization to do that is so powerful. So organization science became, for me this toolkit about kind of like collectively solving problems. I. And then I was, um, hired at Stanford Engineering after I graduated with my PhD. And that was, I mean, that like shaped my career. I feel really lucky. Not every organization science department is an in an engineering school, but in the Stanford engineering I started collaborating with a computer scientist named Michael Bernstein.
[00:02:57] And so sort of from like that moment [00:03:00] forward technology has been a part of every kind of like organizational science question that I've asked. So, um, and it's, I started to see that AI is this other like really powerful tool. And when you think about them together, like organizations plus ai, yeah. The, the hope is that that makes us like even more powerful.
[00:03:18] I think as, as groups, as a society, it's these like really powerful tools for solving problems.
[00:03:23] Nichol: And the way that I think about organizations and AI too is I feel in a period right now where all of human, human society is sort of speed running how we live with ai,
[00:03:36] Melissa: Yeah.
[00:03:37] Nichol: I feel that organizations and AI are some of the first places where there's enough external forces that are forcing.
[00:03:47] Companies to figure it out. But I think that there are also laboratories about how more broadly humans will live with ai.
[00:03:57] Melissa: Yeah, I think that's a really great [00:04:00] perspective. Things that I'm seeing are, so for example, um, some of the studies that I'm doing, I get to watch as people are kind of like experimenting with AI and sort of like reconfiguring their roles at work, which I think is a really powerful thing that all of us will do.
[00:04:14] Kind of like thinking about what does it mean to be human with this powerful technology? So I think the workplace is a great place where that's happening. And then I'm not sure if this is like where you're headed, but I, I think that organizations are also this place where we like negotiate a lot of the social issues of, you know, things like, um, I don't know, like even just like, you know, belonging work and family, um, inclusion, like all of those sorts of things really get negotiated and discussed in organizations.
[00:04:40] Um, and so bringing AI into that is, I think that that conversation is gonna happen in the workplace also.
[00:04:45] Nichol: Mm-hmm. Yeah. Let's talk about flash organizations. For people who don't know what that is, could you explain what they are and then why are they a powerful [00:05:00] model for rethinking how work gets done?
[00:05:03] Melissa: Yeah, so my dissertation, I actually studied teams in the emergency room, and what was so interesting about teams in the emergency department is that you'll have groups of basic, basically they're strangers. They all come together when there's like, you know, a, a patient will come in an ambulance, and then you have this team of strangers that.
[00:05:21] Runup and they all have the expertise that's needed to really like help in that situation. So my dissertation was focused on that. And then when I was hired at Stanford Engineering, um, I met, um, the, as I said, Michael Bernstein, he is a, a professor of computer science and he had been working on all of this research on crowdsourcing.
[00:05:38] Where basically you can use a platform that will let you sort of connect with like millions of people online. But what they were seeing with their crowdsourcing work is that they were having a hard time like building models of crowdsourcing that had a lot of impact. I. What they needed was models of organizing that basically, that were like relative strangers could get together and do something really complex.
[00:05:59] [00:06:00] So you can start to see the connection there. So they were, um, his team was realizing that some of these tools from organization science, where you can understand how temporary groups of strangers can get together and do something really complex, they started seeing that that toolkit was gonna be really useful for crowdsourcing.
[00:06:16] So that was almost 10 years ago that we did the first studies where we were combining these ideas about like temporary teams and temporary organizations and like how to help them be really effective with the crowdsourcing model where use a platform to really like connect with like millions of people.
[00:06:31] So like about 10 years ago we did our, did our first flash teams studies. Um, and since then the platforms have gotten better. Like the data have. Become like even more, you know, prevalent and available. Um, so it's, uh, it's extended to now. It's not just about like temporary teams, it's this whole kind of model of organizing where you can use data basically to pull together a team and then use data to manage the team.
[00:06:57] So that's sort of the arc of it that really takes you to the [00:07:00] frontier now of like what's possible using data.
[00:07:02] Nichol: So looking ahead, how do you see organizational design evolving as AI becomes a more integrated part of a team formation management collaboration? What's going to change?
[00:07:15] Melissa: So with organizational design, right, you'll have somebody who has an idea of something that they want to build. So I can give you one of our kind of inspiring stories as an example. So we worked with, um, a guy who was, he was a medical resident at a hospital and he had this idea of an app that was gonna really solve a problem in his, um, in the place where he worked.
[00:07:36] But he had never built software before. So what he needed was like a lot of help on who he needed to hire, how to sort of structure the team, how to sort of structure their workflow and all of those things. Um, AI can be really helpful with, so, um, different things that AI can be helpful with is, um, first helping him, it helped him like search for experts.
[00:07:57] So for all of our flash teams we did, uh, we [00:08:00] integrated with a platform called Upwork. So Upwork has like millions of contractors. So you can do AI search for experts. So he can like type something in that's like, I wanna build this piece of software. And then AI can recommend to him, here's the experts that we think you're gonna need.
[00:08:17] And then he's never, you know, led a team before. So he doesn't know how to structure the experts, he doesn't know the workflow. But that has been done, you know, hundreds if not millions of times before. So you can have, um, a record of how to structure the team, how to structure the workflow. And then at that point it's just patterned recognition and then kind of a recommendation to him of, you know, so here's the experts, here's how to structure them, here's how to, uh, structure their workflow.
[00:08:42] And in a little bit we could talk about different interventions where you can kind of monitor how things are going and then kind of, you know, do interventions to help things go better. But I think it's like, it's that idea of searching for experts and then using kind of past successful teams to learn how to structure the experts.
[00:08:57] I think that's where AI can really augment [00:09:00] organizational design.
[00:09:01] Nichol: I think also we're entering a period where, you know, from vibe coding with people, designing software with their voice to the ability to use AI to help us plan work. It's a period where. Really, anyone can build something that matters to them.
[00:09:25] Melissa: Yeah.
[00:09:26] Nichol: I think also one of the other things about teams too is that multidisciplinary, um. The, the multidisciplinary aspects of how so many solutions, having like that flash team of expertise and then finally being able to use AI to make it work better. I, I think that's exciting. I.
[00:09:54] Melissa: Yeah, I totally agree. I think that's a really nice thing to emphasize is it's very [00:10:00] interdisciplinary. You need very specialized experts. So it's, it is a really, it's an exciting moment where we collectively can solve more problems together. I like how you said that, and then AI is gonna enhance that collective capability.
[00:10:13] Nichol: So a lot of our audience, they are HR leaders and they are in the middle Their companies adopting AI and their rethinking workflows and also, you know, learning and development, helping their. Employees get up to speed, but they're getting a lot of questions around automation versus augmentation, and I read one of your in research about the fashion buyers using AI to enhance their decisions. you sort of speak to how AI can amplify human expertise, especially for those people in our audience who are getting this [00:11:00] exact same question. their employees.
[00:11:03] Melissa: I love that you read that study. Thank you. I think that's a really, um, that's a great one. Um, it was because there were things about it that were pretty unique and rare, and it's this really, it's a cool story to get to talk about. Um, so it was a study that I did at a company where I got to shadow some data scientists.
[00:11:20] And things that I learned from this study is that the data scientists went in with this explicit frame around augmentation, and there were really cool moments in the design of the algorithmic system where they were like very deliberately kind of ideating, like what were the new capabilities that the buyers would have when they used the tool?
[00:11:41] So there was a pretty dramatic shift in how the buyers used to make their decisions and then that how they started to make decisions using the algorithm. And it was really cool to see the data scientists like frame around new capabilities and they could sort of like anchor the conversation so that the buyers, you know, if something funny is going on with the algorithm, which it always does at the start, [00:12:00] then the buyers could sort of like, you know, look back to that really that anchored frame around augmentation.
[00:12:05] Here's what you'll be able to do better. So I think that, um, the data scientists really deliberately building for and framing around augmentation was, I mean, I think that's a real opportunity. I think that leaders and companies can kind of look to that and can think about, um, how can we have help our experts have more impact.
[00:12:25] You know, we take that as an explicit frame on building tools and, you know, sort of developing these workflows. There was a lot of like re-skilling that had to happen. I could say some about that, but I think that like the key is that augmentation is a choice, like it is a frame. It is a choice that can be made by leaders and then the technology developers.
[00:12:44] Nichol: I read one of your colleagues' papers on, or article on the touring trap
[00:12:50] Melissa: Mm-hmm.
[00:12:50] Nichol: that I think speaks to that as well, saying that You know, there there's one line of of reasoning that people do where they're looking [00:13:00] to imitate and thus replace humans.
[00:13:03] Melissa: Mm-hmm.
[00:13:03] Nichol: another one where it's sort of, how do we augment humans?
[00:13:08] Melissa: Yeah, that's a great paper. Um. Right. There are things, and I think what's interesting about this example, um, in the, the fashion buyers that you're pointing out is there are things that the algorithm just like frankly does better. And like the people can be augmented because the mach like so is an optimization algorithm, right?
[00:13:27] Like it just calculates things that like models things in a way that helps the experts make better decisions. So, um. I think what you're like really rightly pointing out with that example is, um, the design opportunity is to get really clear on what the experts are gonna be needed for things like strategy, things like curating the algorithmic output, stuff like that.
[00:13:48] You definitely need the experts to be helping. But there's also a lot of stuff that they should not be doing anymore, either like the, like the, you know, the analyses that the algorithms can do in such a high value way. Like [00:14:00] you definitely want to like hand that off and then keep the experts doing the high value stuff that they can do.
[00:14:05] Nichol: Where are we? Uh, research and science on determining which tasks are best for automation, which tasks are best for augmentation, uh, augmentation. The reason why I'm asking that is I think a lot of people in our audience and then just out in the world who are new to ai, one, they think it's all one thing,
[00:14:28] Melissa: Right.
[00:14:28] Nichol: two, there really is this, you know, differentiation.
[00:14:33] It sounds like where there are times for. You know, there are times for different things, like I call it, um, you know, it's like just because we now make ketchup does not mean we're not gonna want. Mustard. Sometimes you want mustard, sometimes you want ketchup, and sometimes you want mustard and ketchup and a little bit of mayo. And so I think it's kind of like that where we're [00:15:00] getting to the point where people hopefully begin to understand that it's not all one thing and there's different times that you would want more. Automation in times where you'd want more augmentation. you tell us where we are in sort of like understanding what's best when.
[00:15:18] Melissa: Yeah, I think that's a really great distinction. And I like the metaphor. I think, um, the dimension that I would kind of point leaders to is thinking about the difference between something where you're looking for like precision and accuracy. So then you're gonna want something that's more like machine learning, like, it's more like about predictive analytics.
[00:15:38] Sometimes when people say ai, that's what they're talking about. Um, so I don't know, something like a medical diagnosis, you know, where you have like the scans and like the, um, machine learning algorithm is going to, um, more accurately identify like an an image, for example, compared to a person. So when you really need accuracy and precision.
[00:15:57] That's something that machines can really be [00:16:00] good at in a way that, that people are not. Um, so those kinds of tasks, those kinds of high value tasks around. Like accuracy and precision. Um, have, like that, that kind of thing. It's really useful to have a machine learning system. What's really interesting right now is, um, we're also being asked to like, think about, and when I say we, I mean broadly because, uh, with, uh, generative ai, um, so all of the kind of self-serve generative AI tools, chat, GPT, Gemini, whatever it is, like we are now all kind of figuring out which of our own tasks we can use generative AI on.
[00:16:35] And those are very different value propositions, very different use cases. Um, that is not presently all about like accuracy and precision, right? It's just about, you know, uh, you write your email faster, you write your blog post faster. Um, so I think like differentiating between those two, like that's your catch up, that's your mustard.
[00:16:55] Um, I will say some of the research that I'm doing now with, um, one of my, uh, one [00:17:00] of the companies that I'm doing research with is, um, they're looking to kind of combine how can you combine. The kind of the precision of the more like analytical stuff and the sort of like interactivity. Um, and just like the kind of, you know, content creation interactivity of like a chat GPT.
[00:17:17] So it does seem like there will be, 'cause you were saying catchup and mustard, it does seem like there will be kind of, the frontier might be those sort of like business analytics and that generative ai, like being able to bring those together in a powerful way. That seems like, um, a question that I see a lot of people kind of looking to now.
[00:17:34] Um, I think in terms of like, which tasks for generative ai, I will say, um, some of the most exciting, interesting research that I'm doing right now is, um, getting to like watch as experts and companies really like, they're, they're really like figuring it out. They're like really inventing their own use cases and it's, it's a cool technology in that way.
[00:17:54] It's very different than a machine learning system. It's, you know, 'cause it's so interactive, it's so [00:18:00] experimental. It's so flexible. So this like general purpose technology, we're just like at a moment where you, like everyone's inventing a use case. Like all of my interviews, I'll be like, what did you use it for?
[00:18:10] Like, everyone's got a use case that they themselves came up with, which is, it's a, it's a very, um, different adoption path and it's really exciting.
[00:18:18] Nichol: Yeah. Well, to take that a step further on the use cases, are the design principles that audience should keep in mind when they're thinking about having and designing flexible AI enabled teams? But human centered, like what are, what are the design principles that they should keep in mind and that you advise companies to consider?
[00:18:50] Melissa: Yeah. So when you're thinking about, um, flash teams. As, as kind of a, an inspiring technology. So if [00:19:00] you're, so, for example, even if you're in a really large company and you just like see opportunity to maybe be more innovative, to be more adaptive, um, then. Kind of the principles around thinking more, you know, in like the flash team's mindset, um, we always say it's like the mindset of recognizing that there's like experts everywhere all the time.
[00:19:20] Nichol: Mm.
[00:19:20] Melissa: And just like, sort of like more flexibly thinking about how can I draw together a group of experts to solve this problem, to build, build this new thing. Um, so the, I think the, the key is like the mindset of recognizing experts everywhere all the time. Um, I'll, I'll give credit. My office neighbor is Bob Sutton and he, um, he like, he's really taken to that idea and so he'll like point out all the times that like you're in a flash team, like, to some degree, right?
[00:19:46] To like the creation of this pod podcast. We don't know each other. We all got together and now we have this like, beautiful podcast that we created. So like to some degree, like this is a pod, this is a flash team. So starting to like, just see like what are all the like moments of like [00:20:00] expert collaboration and how can you.
[00:20:02] As, um, like a leader in an organization, how can you like, use that mindset to really become more innovative? Um, I'll say in terms of like the, um, the capabilities, right? So that's the mindset. Um, we talk a lot in our, our book. So we have a book that's coming out on flash teams. We talk a lot about the kind of proof of concept software that you can use for it.
[00:20:23] Um, we were really interested in like, you know, testing and creating software that. Like at, at like in the future, we're like envisioning all these like AI enabled things. Even if you don't have that software, you can just, like, it's things like, um, even just like with a, like with a spreadsheet, you can kind of be like, who, who are the experts who are available and kind of like play around with like a, like a team staffing algorithm, um, where you're like, here's who's available, here's when they're available, uh, here's who's available, here's who's worked together before.
[00:20:53] So you, like, you start to have a point of view of like the team you're trying to build. And like with that, you're starting to have some like [00:21:00] data, some like some team design principles behind it. So it's like, it is very within anybody's, um, kind of reach at this moment to do, to do stuff like this.
[00:21:09] Nichol: Yeah, I had a friend who was in a meeting a few weeks ago, and while they were talking about a internal tool that they wanted to use, someone on the team built the prototype using of the no-code software tools,
[00:21:28] Melissa: Okay.
[00:21:29] Nichol: that meeting turned into a workshop.
[00:21:31] Melissa: Amazing. Yes,
[00:21:32] Nichol: amazing
[00:21:33] Melissa: yes. Yikes. Yes. Yeah.
[00:21:35] Nichol: It's really, really amazing and, and that sort of leads us into the culture around change and the culture around people really sort of. Understanding flash teams that a group of people come together, expertise is everywhere and it's kind of a reorientation around how people think about teams and [00:22:00] teaming. So how do you approach, or how would you recommend our audience approach? Building a culture that can support flash teams, uh, and and temporary tech enabled teams without sacrificing some of the cohesion and trust and sense of purpose that, you know, people have had in the past, in the pre, in the previous team structures.
[00:22:26] Melissa: Yeah, for sure. I think so. I think you're, um, I. You're framing this really important problem, which is, um, so I think the way that I would think about it is you have the big group that the experts can be drawn from. So like, you could, I mean, just like we could use Stanford as an example. So Stanford has like, you know, so many people and, uh, they have full-time jobs.
[00:22:49] They have, you know. Healthcare, they have retirement, they have like career ladders, they have each other. We have like HR, which is really important. So Stanford is this like huge, um, [00:23:00] organization that provides all of those benefits for the community members. Um, and then we all have these like little research projects that we like spin up for like a year, you know, six weeks to a year or to like five years.
[00:23:11] And those, I think it's like the, um, there's different kinds of culture and those two groups are solving different problems. You want like stability, you want benefits, you want community for folks. And then you also, um, like with the little flash team, so like our little, like our collaboration where we write a paper together.
[00:23:27] Sometimes we'll do that in like six weeks. We come together and we write the paper. And I think culture within that, I think it's like a different. It's providing a different benefit for people.
[00:23:36] Nichol: Mm-hmm.
[00:23:37] Melissa: what I really love about flash teams is, um, kind of like the solidarity and the excitement around solving a problem, producing something, being creative together.
[00:23:45] So oftentimes because flash teams are so focused and you're just like building a thing together, the culture and the solidarity can kind of take care of itself. Like it's just people are so excited to build together. And then, but I think the like really important point that you're noticing is, um, there [00:24:00] needs to be a container.
[00:24:01] Around those like moments of like really focused building and really focused solidarity that provides like culture benefits, like all of that great stuff.
[00:24:09] Nichol: So once again, knowing when you're doing, knowing when you need ketchup, knowing when you need mustard. Um, and it's sort of like this thing about, you know, having the, the time and the context. Um, and really I think also communicating that so people understand that the flash, he is a framework for a specific goal.
[00:24:32] Melissa: Mm-hmm. Yeah, exactly. Yeah.
[00:24:34] Nichol: So what surprised you? Like what's a surprising lesson that you've had on your work with flash organizations and AI supported collaboration, especially with leadership and accountability? Was there something that you did not expect that you discovered?
[00:24:54] Melissa: Um, I mean, I have to be honest, I feel like, uh, generative AI took me and many people by surprise. [00:25:00] Like in terms of like how much that, um. Really shifted the conversation and the capabilities. Um, so because we've been, you know, building flash teams and kind of doing this research for a long time, and then really it's, I mean, you just gave the example like now somebody in a meeting can, you know, go into whatever tool, chat GBT and be like, you know, code me a website.
[00:25:22] We used to be building websites using teams of people, and now like that website is like done, you know, in a snap. Um, so I think the thing that has truly surprised me the most is this shift in capabilities. We are all so much more capable. Augmented with, these gen AI tools. and that's true for flash teams also.
[00:25:41] I think we are all at this moment of figuring out which tasks do I hold on to? Which tasks do I delegate, within our own jobs? and I'm seeing people look more to the team question there. you and I are probably both using these tools. you get into a rhythm with yourself, but [00:26:00] what does it look like on a team?
[00:26:01] when a flash team, when everybody has like a copilot, right? what does that look like? So that has really been, um, interesting to think about. I don't know that I have a lot of answers, but that has been a surprise for sure.
[00:26:11] Nichol: Yeah, the, so my work has been focused on human potential and technology for a decade, and a big part of that is mental, emotional, and social health. and the pandemic made people, I. Interested in mental health and then remote work made people interested in social health. But the thing about generative AI that I never imagined was that made people interested in. Understanding what it means to be human, which made them then interested in emotional health.
[00:26:49] Melissa: Ooh,
[00:26:49] Nichol: if you had, if you had asked me a decade ago a or if you had said, Hey Nicole, AI's gotta make people care about emotional health, I would've like, [00:27:00] it just never would've crossed my mind. So
[00:27:01] Melissa: I totally agree. Yes, yes, yes.
[00:27:05] Nichol: Okay. So, uh, do we measure. When we're successful, when we have AI on teams and, and with new organizational structures, like how do we know that it's working and, and how do we measure that to know it's truly supporting people in performance?
[00:27:25] Melissa: Hmm. Okay, so that's a great question and I'm gonna have to go for the ketchup and mustard again by say. Right, because we, um, we've been doing flash teams for a long time, and I mean more so in the, like, analytical, like the more so in like the business insights way of thinking about performance, right? Did you meet a deadline?
[00:27:49] Um, you know, did you meet your milestone, whatever it was? Um, even within that world, you could ask questions that are, um, more kind of. Socially focused like team [00:28:00] cohesion. Do you wanna work together again? Um, so, so that's, um, like, it's a way of saying that that flash teams sort of produced quite a lot of data that allows you to ask really interesting questions about performance.
[00:28:14] And that's where we were thinking for a long time. And it was like, it was quite, I think it was quite interesting. I do feel like the frontier has shifted. I mean, as you and I are both sort of acknowledging our own, like this is a moment to really be thinking like, what's, what's new, what's different? So I think things have, um, I think that that capability for measurement is really powerful and, uh, what I'm seeing now, there's so much, like, so much is shifting with these like generative AI tools.
[00:28:40] Um, so far I see more individual adoption. Um, I'm like one of the studies that I'm doing right now. We, we very purposefully structure the study to look at managers and kind of whole teams as they're adopting gen AI to like be able to ask the relationship questions. How does individual adoption change, like [00:29:00] what it means to be a teammate?
[00:29:01] What does it, how does it change your relationship with your manager? What happens if you're managing an ai? You know, you're like managing an agent. Um, so we're trying to ask the relational questions that does feel to me to be the, like, research frontier. Um, there was a study that came out of, um. Uh, Wharton and Harvard.
[00:29:17] I think probably last week you probably saw this one. It was like the first time that I saw a published study that had, uh, teams using AI rather than individuals using ai. Um, and they did find that the teams using ai, um, produced some of the very best solutions. It was like a new product development task, I think.
[00:29:36] Um, so they ended up being, the teams plus the AI were like the most. Uh, innovative, which I thought was really cool. Uh, that's one study. So anyways, the ketchup and mustard there, right, is like all the data on the team processes. And then now, you know, the frontier question is how does generative AI change all of those relationships?
[00:29:54] And I think that's like the fun thing to be studying right now.
[00:29:56] Nichol: Yeah, I agree. I'm, I'm actually obsessed [00:30:00] about the team part of it because I think that collaboration at scale. Real collaboration because I don't count Facebook
[00:30:09] Melissa: Great.
[00:30:10] Nichol: collaboration, but like real human solving real problems collectively. I think that's our, you know, human sup. What it's, uh, collective human super intelligence.
[00:30:24] Melissa: totally.
[00:30:24] Nichol: how we get there.
[00:30:25] Melissa: Yeah.
[00:30:26] Nichol: so I think the small work gets done on teams and organizations learning from that and then understanding how those, Insights can scale to, you know, larger and larger groups of humans.
[00:30:42] Melissa: Yeah, absolutely.
[00:30:44] Nichol: Yeah,
[00:30:45] Melissa: I think that's the vision. I love that description.
[00:30:47] Nichol: Yeah. Yeah. Okay. So our audience, when after they listen to our conversation, are the practical steps that you [00:31:00] would recommend to them?
[00:31:02] For those who are just starting their AI journey, they, as soon as they listen to our podcast and then they get to work, what do you suggest that they do? Where should they start?
[00:31:12] Melissa: It seems that for most people, uh, generative AI is top of mind. So I'll speak first to what I'm seeing practically with people like learning to adopt gen ai. Um, and then I will say how I think, uh, flash teams can actually augment that journey. So, um, I think in terms for HR leaders, um, what I'm seeing, um, so the fun thing about my job and the way that I do research is I get to really have like a front row seat to adoption in companies right now.
[00:31:42] Um, and it, it. I think pretty consistently. I see. Um, that like executives, you know, so probably like your, um, chief HR officer is like bought in your CEO is bought in. Everybody's like, this is happening, this is cool. And then like really broadly, I'm not seeing a lot of like [00:32:00] individual like resistance and the frontline people are experimenting, like, kinda like everybody's using it.
[00:32:05] So like every, like all the executives are like, we're doing this and everybody's using it. Um, and it seems like the opportunity is really, as you and I are saying, like right at that like team level. Like how, how does this affect teams? How does it affect departments? Um, so I think that's the opportunity.
[00:32:19] That's where I'm seeing a lot of, um, sort of like opportunity for leadership, for framing, for figuring out what to build. So I like, I would encourage people to be looking at that, that middle layer, um, between individual experimentation and kind of top down vision. Um, so I think that's for Gen ai. Um, and then if somebody is like, excited about Flash teams, um, and like what, how do you kind of get started with that? I mean, I think that the really exciting opportunity with flash teams is this perspective that we're seeing of it's at a moment where people are afraid of replacement and job loss. But I think if we focus, refocus on collectively, we solve problems. And if you unfreeze [00:33:00] the organizational system and say, we have been solving these problems. They're changing because of ai. What are the new problems we can solve? And if you open that culture and frame that out for people, sort of vision of it is refocusing people on solving new problems in innovative ways.
[00:33:18] Nichol: Mm-hmm. Absolutely. The, when I think about it, it's sort of like the, you know, the, the places that. there's so much room to move into solving new problems, finding new problems. So expanding the problem space, finding new problems to solve, finding new ways to solve those problems. And then also it's sort of what I think of as being, especially in organizations, being an entrepreneur. So, you know, when you're an entrepreneur, there is always more. To do. There are always
[00:33:54] Melissa: Yeah.
[00:33:55] Nichol: ways to serve more customers. And so, you know, [00:34:00] with both of those, those styles unlocked, there is so much more to do. Like
[00:34:05] Melissa: Yes,
[00:34:05] Nichol: we don't have our, you know, we, we don't have our flying cars yet. You know, like there is so much to do. There's so many new problems that we could
[00:34:14] Melissa: yes.
[00:34:15] Nichol: and um, I agree with you. I think flash teams are. and just in general how teams use AI is a big way of being able to expand into that possibility space because like that's where the new jobs are.
[00:34:31] Melissa: Yeah, totally.
[00:34:32] Nichol: that nobody knows where they are yet. That's where they are.
[00:34:36] Melissa: Yes.
[00:34:36] Nichol: so we have to like, we have to do this. Oh gosh. Well I could talk to you forever.
[00:34:42] Melissa: Yeah.
[00:34:42] Nichol: Um, but that's it for this week's episode.
[00:34:46] Melissa: sounds
[00:34:46] Nichol: A big thank you to you, Melissa Valentine, for sharing your experience and insights with us. And everyone. Before we wrap up, I encourage you to follow the A IHI project wherever you enjoy your [00:35:00] podcast. And if you enjoyed today's episode, please take a moment to comment, leave a review, and help spread the word. And finally, you can find all of our episodes on our website at SHRM dot org slash ihi. Thanks for joining the conversation and we will catch you next time.
[00:35:18]
[00:35:25] Brightmine introduces AI assist a new gen AI powered tool in the HR and compliance center, providing HR professionals with swift and reliable answers to complex compliance questions.
[00:35:36] Request a quote at brightmine.com.
[00:35:40]